windelband, history of philosophy, vol. 1

398

Click here to load reader

Upload: maivin2

Post on 28-Oct-2014

233 views

Category:

Documents


31 download

TRANSCRIPT

Page 1: Windelband, History of Philosophy, Vol. 1

f

Page 2: Windelband, History of Philosophy, Vol. 1

UNIVERSITYOF FLORIDALIBRARIES

Page 3: Windelband, History of Philosophy, Vol. 1
Page 4: Windelband, History of Philosophy, Vol. 1

Digitized by the Internet Archive

in 2010 with funding from

Lyrasis Members and Sloan Foundation

http://www.archive.org/details/historyofphiloso01wind

Page 5: Windelband, History of Philosophy, Vol. 1
Page 6: Windelband, History of Philosophy, Vol. 1
Page 7: Windelband, History of Philosophy, Vol. 1

HARPER TORCHBOOKS / The Cloister Library

Tor AndraeAugustine/ PrzywaraRoland IT. Bainton

Karl BarthKarl Barth

Nicolas BerdyaevNicolas Berdyaev

James Henry Breasted

Martin Buber

Martin BuberJacob Burckhardt

Edward ConzeF. M. Cornford

G. G. CoultonG. G. Coulton

H. G. CreelAdolf Deissmann

C. H. DoddJohannes Eckhart

Mircea Eliade

Morton S. EnslinMorton S. Enslin

Austin Farrer, ed.

Ludwig FeuerbachHarry Emerson Fosdick

Sigmund Freud

Maurice Friedmanctavius Brooks Frothingham

Edward Gibbon

Edward Gibbon

Charles C. Gillispie

Maurice Goguel

Maurice Goguel

Edgar J. GoodspeedHerbert J. C. Grierson

William HallerAdolf HarnackEdwin Hatch

Karl HeimF. H. Heinemann

Stanley R. Hopper, ed.

Johan HuizingaAldous Huxley

S0ren Kierkegaard

mohammed: The Man and His Failh tb/62AN AUGUSTINE SYNTHESIS TB/35THE TRAVAIL OF RELIGIOUS LIBERTY TB/30DOGMATICS IN OUTLINE TB/56THE WORD OF GOD AND THE WORD OF MAN TB/ I 3

THE BEGINNING AND THE END Tls/14

THE DESTINY OF MAN Tß/6lDEVELOPMENT OF RELIGION AND THOUGHT IN ANCIENT

EGYPT TB/57eclipse of god : Studies in the Relation Between Religion

and Philosophy TB/12moses : The Revelation and the Covenant TB/27THE CIVILIZATION OF THE RENAISSANCE IN ITALY [Illus-

trated Edition] : Vol. I, TB/40 ; Vol. II, tb/4ibuddhism : Its Essence and Development TB/58from religion to philosophy : A Study in the Origins

of Western Speculation TB/20MEDIEVAL FAITH AND SYMBOLISM TB/25THE FATE OF MEDIEVAL ART IN THE RENAISSANCE AND

REFORMATION TB/26CONFUCIUS AND THE CHINESE WAY TB/63paul: A Study in Social and Religious History TB/15THE AUTHORITY OF THE BIELE TB/43meister eckhart : A Modern Translation tb/8cosmos and history : The Myth of the Eternal Return

TB/50christian beginnings tb/sthe literature of the christian movement tb/6THE CORE OF THE BIBLE Tb/7THE ESSENCE OF CHRISTIANITY Tb/iIA GUIDE TO UNDERSTANDING THE BIBLE Tb/2on creativity and the unconscious : Papers on the

Psychology of Art, Literature. Love, Religion TB/45martin buber : The Life of Dialogue TB/64TRANSCENDENTALISM IN NEW ENGLAND: A History

TB/59THE END OF THE ROMAN EMPIRE IN THE WEST [J. B. Bury

Edition ; Illustrated] TB/37THE TRIUMPH OF CHRISTENDOM IN THE ROMAN EMPIRE

[J. B. Bury Edition ; Illustrated] TB/46GENESIS AND GEOLOGY TB/51jesus and the origins of Christianity i : Prolegom-

ena to the Life of Jesus tb/65JESUS AND THE ORIGINS OF CHRISTIANITY II : TIlC Life

of Jesus tb/66A LIFE OF JESUS Tb/iCROSS-CURRENTS IN 17TH CENTURY ENGLISH LITERATURE:

The World, the Flesh, the Spirit TB/47THE RISE OF PURITANISM Tb/22WHAT IS CHRISTIANITY? TB/17THE INFLUENCE OF GREEK IDEAS ON CHRISTIANITY

TB/18CHRISTIAN FAITH AND NATURAL SCIENCE TB,/l6

EXISTENTIALISM AND THE MODERN PREDICAMENT TB/2SSPIRITUAL PROBLEMS IN CONTEMPORARY LITERATURE

TB/21ERASMUS AND THE AGE OF REFORMATION TB,/lO

the devils of loudun : A Study in the Psychology of

Pozver Politics and Mystical Religion in the France ofCardinal Richelieu tb/6o

edifying discourses: A Selection TB/32(Continued on next page)

Page 8: Windelband, History of Philosophy, Vol. 1

S0ren KierkegaardS0ren KierkegaardAlexandre Koyre

Emile Male

H. Richard NiebuhrH. Richard Niebuhr

H. J. RoseJosiah Royce

. Auguste Sabatier

George SantayanaGeorge Santayana

Friedrich SchleiermacherHenry Osborn Taylor

Paul Tillich

Edward Burnett TylorEdward Burnett Tylor

Evelyn UnderbillJohannes Weiss

Wilhelm Windelband

Wilhelm Windelband

THE JOURNALS OF KIERKEGAARD TB/S2PURITY OF HEART TB/4FROM THE CLOSED WORLD TO THE INFINITE UNIVERSE

TB/31the gothic image : Religious Art in France of the 13t)

Century TB/44CHRIST AND CULTURE TB/3THE KINGDOM OF GOD IN AMERICA TB/4QRELIGION IN GREECE AND ROME TB/55the religious aspect of philosophy : A Critique 0,<

the Bases of Conduct and of Faith tb/sooutlines of a philosophy of religion based on psy-

chology AND HISTORY TB/23INTERPRETATIONS of poetry and religion TB/9winds of doctrine and

PLATONISM AND THE SPIRITUAL LIFE TB/24on religion : Speeches to Its Cultured Despisers tb/3(

THE EMERGENCE OF CHRISTIAN CULTURE IN THE WESTThe Classical Heritage of the Middle Ages TB/48

DYNAMICS OF FAITH TB/42THE ORIGINS OF CULTURE TB/33RELIGION IN PRIMITIVE CULTURE TB/34WORSHIP TB/lOearliest Christianity : A History of the Period a.di

30-150: Vol. I, TB/53 ; Vol. II, TB/54a history of philosophy i: Greek, Roman, Medieva,

TB/38.

a history of philosophy 11 : Renaissance , Enlighten]ment, Modern TB/39

HARPER TORCHBOOKS / The Academy Library

H. J. BlackhamWalter Bromberg

G. P. Gooch

Francis J. GrundHenry James

Georges Poulet

six existentialist thinkers TB/1002the mind of man : A History of Psychotherapy anc

Psychoanalysis TB/1003ENGLISH DEMOCRATIC IDEAS IN THE SEVENTEENTH CEN-!

TURY TB/1006ARISTOCRACY IN AMERICA TB/lOOITHE PRINCESS CASAMASSIMA TB/lOOSSTUDIES IN HUMAN TIME TB/1OO4

HARPER TORCHBOOKS / The Science Library

J. BronowskiW. C. Dampier, ed.

Arthur Eddington

H. T. Pledge

George SartonPaul A. Schilpp, ed.

Friedrich Waismann

W. H. Watson

G. J. Whitrow

A. Wolf

SCIENCE AND HUMAN VALUES TB/505READINGS IN THE LITERATURE OF SCIENCE TB/SI2space, time and gravitation : An Outline of the Gen-

eral Relativity Theory tb/sioscience since 1500: A Short History of Mathematics,

Physics, Chemistry, and Biology TB/506ANCIENT SCIENCE AND MODERN CIVILIZATION TB/5OIalbert einstein: Philosopher-Scientist : Vol. I,

TB/502 ; Vol. II, TB/503introduction to mathematical thinking: The Forma-

tion of Concepts in Modern Mathematics tb/siion understanding physics : An Analysis of the Philos-

ophy of Physics TB/507THE STRUCTURE AND EVOLUTION OF THE UNIVERSE : An

Introduction to Cosmology TB/504A HISTORY OF SCIENCE, TECHNOLOGY AND PHILOSOPHY INthe i6th and 17TH centuries: Vol. I, TB/508 ; Vol. II,

TB/509

Page 9: Windelband, History of Philosophy, Vol. 1

AHISTORY

OFPHILOSOPHY

VOLUME I

Greek, Roman, and Medieval

WILHELMWINDELBAND

HARPER TORCHBOOKS

HARPER & BROTHERS, PUBLISHERSNew York

Page 10: Windelband, History of Philosophy, Vol. 1

A history of philosophy : I—Greek, Roman, and Medieval

Printed in the United States of America

Reprinted by arrangement with The Macmillan Company, New York, fromthe revised edition of 1901, translated by James H. Tufts.

First Habpeh Tohciibook edition published 1958

/9SZ

Library of Congress catalog card number: LC-58-7114—Vol. I

Page 11: Windelband, History of Philosophy, Vol. 1

TRANSLATOR'S PREFACE.

oXKc

Regarded simply as a historical discipline, the history of thought

might fairly claim a prominent place in education, and an equal

share of the attention now given to comparative and historical

studies. The evolution of an idea is in itself as interesting and

valuable an object of study as the evolution of a word, of an insti-

tution, of a state, or of a vegetable or animal form.

But aside from this interest which it has in common with other

historical sciences, the history of philosophy has a peculiar value of

its own. For the moment we attempt any serious thinking in any

field,— natural science, history, literature, ethics, theology, or any

other, — we find ourselves at the outset quite at the mercy of the

words and ideas which form at once our intellectual atmosphere

and the instruments with which we must work. We cannot speak,

for example, of mind or matter, of cause or force, of species or indi-

vidual, of universe or God, of freedom or necessity, of substance or

evolution, of science or law, of good or true or real, without involv-

ing a host of assumptions. And the assumptions are there, even

though we may be unconscious of them, or ignore them in an effort

to dispense with metaphysics. To dispense with these conceptions

is impossible. Our only recourse, if we would not beg our questions

in advance, or remain in unconscious bondage to the instruments of

our thought, or be slaves to the thinking of the past generations

that have forged out our ideas for us, is to " criticise our categories."

And one of the most important, if not the only successful, means to

this end is a study of the origin and development of these categories.

We can free ourselves from the past only by mastering it. Wemay not hope to see beyond Aristotle or Kant until we have stood

Page 12: Windelband, History of Philosophy, Vol. 1

vi Translator 's Preface.

on their shoulders. We study the history of philosophy, not so

much to learn what other men have thought, as to learn to think.

For an adequate study of the history of thought, the main requi-

sites are a careful study of the works of the great thinkers— a

requisite that need not be enlarged on here, although such study is

a comparatively recent matter in both Britain and America, with a

few notable exceptions— and a text-book to aid us in singling out

the important problems, tracing their development, disentangling

their complications, and sifting out what is of permanent value. To

meet this second need is the especial aim of the present work, and,

with all the excellencies of the three chief manuals already in use,

it can scarcely be questioned that the need is a real one. Those

acquainted with the work here translated (W. Windelband's Ge-

schichte der Philosophie, Freiburg i. B., 1892) have no hesitation in

thinking that it is an extremely valuable contribution toward just

this end. The originality of its conception and treatment awaken

an interest that is greater in proportion to the reader's acquaintance

with other works on the subject. The author shows not only

historical learning and vision, but philosophical insight ; and in his

hands the comparative treatment of the history of thought proves as

suggestive and fruitful as the same method applied to other subjects

in recent times. A work like the present could only have been

written with some such preparation as has come in this case from

the previous treatment of Greek and Modern Philosophy at greater

length, and in presenting it to English readers I am confident that

it will meet the wants, not only of special students of philosophy,

but also of all who wish to understand the development of thought.

Teachers will, I think, find it very valuable in connection with

lecture courses.

As regards the work of the Translator, little need be said. He

has tried— like many others— to make a faithful translation into

intelligible English, and is fully conscious that it has been with

varying success. Of course translation in the strict sense is often

impossible, and I cannot hope to have adopted the happiest com-

promise or found the most felicitous rendering in all cases.

" Being " (spelled with a capital) is used for " Sein." Where the

German " Form " seemed to differ enough from the ordinary English

Page 13: Windelband, History of Philosophy, Vol. 1

Translator's Preface. vii

sense of the word to make "form" misleading, I have spelled it

"Form," and the same course has been taken with " Real^ "Re-

alität," where the German seemed to desire to distinguish them from

"wirklich" which has been translated sometimes by "real," some-

times by " actual." " Vorstellung " is usually rendered by " idea,"

following Locke's usage, except in connection with the system of

Leibniz, where "representation " is necessary to bring out his thought.

"Idee," in the Platonic and Kantian use, is rendered "Idea" (spelled

with a capital). The convenient word "Geschehen" has no exact

counterpart, and has been variously rendered, most frequently per-

haps by " cosmic processes." In the additions made to the bibliog-

raphy, no attempt has been made to be exhaustive ; I have simply

tried to indicate some works that might aid the student. It is

scarcely necessary to say that any corrections or suggestions will

be gratefully received and utilised if possible. Material in square

brackets is added by the translator.

In conclusion, I desire to express my indebtedness to my col-

leagues, Professors Shorey, Strong, and Cutting, and Dr. Schwill

for helpful suggestions. My chief indebtedness, however, is to the

critical taste and unwearied assistance of my wife. If I have in

any degree succeeded in avoiding German idioms, it is largely due

to her.

JAMES H. TUFTS.University of Chicago,

July, 1893.

TRANSLATOR'S NOTE TO THE SECOND EDITION.

In preparing this second edition all changes made by the author

in the second German edition have been incorporated either in the

text or in the appendix at the close. In addition, I have included a

brief notice (pp. 663-670) of certain aspects of recent English

thought, which naturally have more interest for the readers of

this translation than for those of the original.

JAMES H. TUFTS.University of Chicago,

May, 1901.

Page 14: Windelband, History of Philosophy, Vol. 1
Page 15: Windelband, History of Philosophy, Vol. 1

AUTHOR'S PREFACE.

=>J*iC

After many painful delays and interruptions I now present at

last the conclusion of the work whose first sheets appeared two

years ago.

The reader will not confuse this with the corapendiums which

have very likely sometimes been prepared by dressing out lecture

notes on the general history of philosophy. What I offer is a

serious text-book, which is intended to portray in comprehensive

and compressed exposition the evolution of the ideas of European

philosophy, with the aim of showing through what motives the

principles, by which we to-day scientifically conceive and judge

the universe and human life, have been brought to consciousness

and developed in the course of the movements of history.

This end has determined the whole form of the book. The

literary-historical basis of research, the biographical and biblio-

graphical material, were on this account necessarily restricted to

the smallest space and limited to a selection that should open the

way to the best sources for the reader desiring to work farther.

The philosophers' own expositions, too, have been referred to in the

main, only where they afford a permanently valuable formulation

or rationale of thoughts. Aside from this there is only an occa-

sional citation of passages on which the author supports an inter-

pretation differing from that ordinarily adopted. The choice of

material has fallen everywhere on what individual thinkers have

produced that was new and fruitful, while purely individual turns

of thought, which may indeed be a welcome object for learned

research, but afford no philosophical interest, have found at most

a brief mention.

Page 16: Windelband, History of Philosophy, Vol. 1

x Author's Preface.

As is shown even by the external form of the exposition, chief

emphasis has been laid upon the development of what is weightiest

from a philosophical standpoint : the history of problems and concep-

tions. To understand this as a connected and interrelated whole

has been my chief purpose. The historical interweaving of the

various lines of thought, out of which our theory of the world and

life has grown, forms the especial object of my work, and I am

convinced that this problem is to be solved, not by any a priori

logical construction, but only by an all-sided, unprejudiced investi-

gation of the facts. If in this exposition a relatively large part

of the whole seems to be devoted to antiquity, this rests upon the

conviction that for a historical understanding of our intellectual

existence, the forging out of the conceptions which the Greek mind

wrested from the concrete reality found in Nature and human life,

is more important than all that has since been thought— the

Kantian philosophy excepted.

The task thus set required, however, a renunciation which no

one can regret more than myself. The purely topical treatment

of the historical movement of philosophy did not permit of giving

to the personality of the philosophers an impressiveness corre-

sponding to their true worth. This could only be touched upon

where it becomes efficient as a causal factor in the combination and

transformation of ideas. The aesthetic fascination which dwells in

the individual nature of the great agents of the movement, and

Avhich lends its especial charm to the academic lecture, as well as

to the more extended exposition of the history of philosophy, had

to be given up here in favour of a better insight into the pragmatic

necessity of the mental process.

Finally, I desire to express at this place also my lively gratitude

to my colleague, Dr. Hensel, who has not only aided me with a

part of the proofs, but has also essentially increased the usefulness

of the book by a subject index.

WILHELM WINDELBAND.

Strassburg, November, 1891.

Page 17: Windelband, History of Philosophy, Vol. 1

AUTHOR'S PREFACE TO THE SECOND EDITION

A large edition of my History of Philosophy had been exhausted

more than two years ago, and in the meantime its use had been

further extended by English and Russian translations. This per-

mits me to assume that the new treatment which I gave to the

subject has filled an existing gap, and that the synoptical and criti-

cal method which I introduced has gained approval so far as the

principle is concerned. While therefore I could leave the book

unchanged in its main outlines when preparing this new edition, I

could be all the more careful in making evidently needed improve-

ments and in fulfilling certain specific requests.

Under the head of improvements I have undertaken such correc-

tions, condensations, and expansions upon particular points as are

requisite for a text-book which seeks to represent the present condi-

tion of investigation, and in this work the literature which has

appeared since the first edition has been utilised. In consequence

of the great condensation of material the exposition had become

sometimes difficult to follow, and 1 have aimed in many cases to

give more fluent form to the expression by breaking up some of the

longer sentences, and occasionally omitting what was of merely sec-

ondary importance.

A desire has been expressed by readers of the book for a more

extended notice of the personalities and personal relations of the

philosophers. In the preface to my first edition I had myself

recognised the justice of this demand, but had disclaimed the inten-

tion of satisfying it because the special plan of my work and the

necessary limitations of space prevented. Now I have sought to

fulfil this demand so far as it has seemed possible within the

limit of my work, by giving brief and precise characterisations of

the most important thinkers.

A desire for a more extended treatment of the philosophers of the

nineteenth century has also been reckoned with. The few pages

originally accorded to the subject have been expanded to three times

the former compass, and I hope that although one will miss one

Page 18: Windelband, History of Philosophy, Vol. 1

xii Author s Preface.

topic and another another, it will nevertheless be possible to gain a

fairly complete general view of the movements of philosophy downto the more immediate present, in so far as this is to be expected

from a history of principles.

Finally, I have remade the subject index, and so expanded it

that in connection with the text it may, as I hope, have the value of

a dictionary of the history of philosophy. This gives to my work a

second distinctive feature; namely, that of a work of reference of

a systematic and critical sort.

By all these expansions the size of the book has been considerably

increased, and I express here to my esteemed publisher, Dr. Siebeck,

my heartiest gratitude for the cordial response with which he has

made possible these essential improvements.

WILHELM WINDELBAND.

Strassburg, September, 1900.

Page 19: Windelband, History of Philosophy, Vol. 1

CONTENTS.

INTRODUCTION.PAOK

§ 1. Name and Conception of Philosophy 1

§ 2. The History of Philosophy 8

§ 3. Division of Philosophy and of its History 18

PART I.

THE PHILOSOPHY OF THE GREEKS.Introduction 23

Chapter I. The Cosmological Period 27

§ 4. Conceptions of Being 31

§ 5. Conceptions of the Cosmic Processes or Becoming ... 47

§ 6. Conceptions of Cognition 57

Chapter II. The Anthropological Period 66

§ 7. The Problem of Morality 72

§ 8. The Problem of Science 87

Chapter III. The Systematic Period 99

§ 9. Metaphysics grounded anew by Epistemology and Ethics . 104

§ 10. The System of Materialism 109

§11. The System of Idealism 116

§ 12. The Aristotelian Logic 132

§ 13. The System of Development 139

PART II.

THE HELLENISTIC-ROMAN PHILOSOPHY.

Introduction 155

Chapter I. The Ethical Period 159

§ 14. The Ideal of the Wise Man 163

§ 15. Mechanism and Teleology 17S

§ 16. The Freedom of the Will and the Perfection of the Uni-

verse 190

§ 17. The Criteria of Truth 197

xiii

Page 20: Windelband, History of Philosophy, Vol. 1

xiv Contents.

PAGE

Chapter II. The Religious Period 210

§ 18. Authority and Revelation 219

§ 19. Spirit and Matter 229

§ 20. God and the World 235

§ 21. The Problem of the World's History 255

PART III.

THE PHILOSOPHY OF THE MIDDLE AGES.

Introduction 263

Chapter I. First Period . . . 270

§ 22. The Metaphysics of Inner Experience 276

§ 23. The Controversy over Universals 287

§ 24. The Dualism of Body and Soul 301

Chapter II. Second Period 310

§ 25. The Realm of Nature and the Realm of Grace .... 318

§ 26. The Primacy of the Will or of the InteUect 328

§ 27. The Problem of Individuality .337

Page 21: Windelband, History of Philosophy, Vol. 1

HISTORY OF PHILOSOPHY.°»Cc

INTRODUCTION.

§ 1. The Name and Conception of Philosophy.

R. Haym, Art. Philosophie in Ersch und Grüber's Encyclopädie, III. Abth.,

Bd. 24.

W. Windelband, Praeludien (Freiburg i. B., 1884), 1 ff.

[A. Seth, Art. Philosophy in Enc. Brit.~]

[G. T. Ladd, Introduction to Philosophy. N.Y. 1891.]

By philosophy present usage understands the scientific treatment

of the general questions relating to the universe and human life.

Individual philosophers, according to the presuppositions with

which they have entered upon their work, and the results which

they have reached in it, have sought to change this indefinite idea

common to all, into more precise definitions, 1 which in part diverge

so widely that the common element in the conception of the science

may seem lost. But even the more general meaning given above is

itself a limitation and transformation of the original significance

which the Greeks. connected with the name philosophy,— a limita-

tion and transformation brought about by the whole course of the in-

tellectual and spiritual life of the West, and following along with

the same.

1. While in the first appearance in literature 2 of the words

<£iAoo-o<£etv and faXoo-o&a the simple and at the same time indefinite

meaning, " striving after wisdom," may still be recognised, the word

" philosophy " in the literature after Socrates, particularly in the

school of Plato and Aristotle, acquired the fixed significance accord-

1 Cited in detail in TJeberwes-Heinze. Grundriss der Gpschichte der Philoso-

phie, I. § 1. [Eng. trans. Ueberweg's History of Philosophy, trans, by G. S.

Morris. N.Y. 1871.]2 Herodotus, I. 30 and 50 ; Thucydides, II. 40 ; and frequently also even in

Tlato, e.g. Apol. 29 ; Lysis, 218 A ; Symp. 202 E ff.

1

Page 22: Windelband, History of Philosophy, Vol. 1

2 Introduction.

ing to which it denotes exactly the same as the German word" Wissenschaft." 1 According to this meaning philosophy in general 2

is the methodical work of thought, through which we are to know

that which " is " ; individual " philosophies " are the particular sci-

ences in which individual realms of the existent are to be investi-

gated and known. 3

With this first theoretical meaning of the word " philosophy " a

second was very early associated. The development of Greek

philosophy came at the time when the naive religious and ethical

consciousness was in process of disintegration. This not only

made the questions as to man's vocation and tasks more and more

important for scientific investigation (cf. below, Part I. ch. 2), but

also made instruction in the right conduct of life appear as an

essential aim, and finally as the main content of philosophy or

science. Thus philosophy in the Hellenistic period received the

practical meaning of an art of life, based upon scientific principles,*—a meaning for which the way had already been prepared by the

Sophists and Socrates.

In consequence of this change, purely theoretical interest passed

over to the particular " philosophies," which now in part assumed

the names of their special subjects of research, historical or belong-

ing to natural science, while mathematics and medicine kept all the

more rigorously that independence which they had possessed from

the beginning with relation to science in general.5 The name of

philosophy, however, remained attached to those scientific efforts

which hoped to win from the most general results of human knowl-

edge a conviction for the direction of life, and which finally culmi-

nated in the attempt (made by Neo-Platonism) to create from such

a philosophy a new religion to replace the old that had been lost.6

1 A conception which it is well known is of much greater compass than theEnglish and French " science." [In this translation the words " science " and" scientific " are used in this larger sense. The term " natural science " will beused for the narrower meaning which "science " alone often has. If it shouldserve to remind the beginner that philosophy and scientific thought should beone, and that natural science is not all of science, it may be of value.]

2 Plato, Rep. 480 B ; Aristotle, Met. VI. 1, 1026 a 18.8 Plato, Thecet. 143 D. Aristotle sets the doctrine "of Being as such" (the

later so-called Metaphysics) as '' First Philosophy '

' over against the other

"philosophies," and distinguishes further theoretical and practical "philoso-phy." In one passage {Met. I. 6, 987 a 29) he applies the plural <pL\o<ro<plai also

to the different systems of science which have followed in historical succession,

us we should speak of the philosophies of Kant, Fichte, Hegel, etc.4 Cf. the definition of Epicurus in Sex':. Emp. Adv. Math. XI. 169, and on

the other hand that of Seneca, JEpist. 89.5 Cf. below, Part I.

Hence Proclus, for example, would prefer to have philosophy called

\heology.

Page 23: Windelband, History of Philosophy, Vol. 1

§ 1.] Name and Conception of Philosophy. 3

There was at first little change in these relations, when the remains

of ancient science passed over into the culture of the present peoples

of Europe as the determining forces of their intellectual life. Con-

tent and task of that which the Middle Ages called philosophy coin-

cided with the conception held by later antiquity. 1 And yet the

meaning of philosophy underwent an essential change by finding

philosophy's task already performed, in a certain sense, by religion.

For religion, too, afforded not only a sure conviction as a rule for

the guidance of personal life, but also in connection with this, a gen-

eral theoretical view of all reality, which was the more philosophical

in its character, as the dogmas of Christianity had been formulated

entirely under the influence of ancient philosophy. Under these

circumstances, during the unbroken dominance of Church doctrine,

there remained for philosophy, for the most part, only the position

of a handmaid to ground, develop, and defend dogma scientifically.

But just by this means philosophy came into a certain opposition to

theology as regards method ; for what the latter taught on the

ground of divine revelation, the former was to win and set forth by

means of human knowledge. 2

But the infallible consequence of this relation was, that the freer

individual thinking became in its relation to the Church, the more

independently philosophy began the solution of the problem which

she had in common with religion ; from presentation and defence of

doctrine she passed to its criticism, and finally, in complete inde-

pendence of religious interests, sought to derive her teaching from

the sources which she thought she possessed in the "natural light"

of human reason and experience.3 The opposition to theology, as

regards methods, grew in this way to an opposition in the subject

matter, and modern philosophy as " world-wisdom " set itself over

against Church dogma.4 However manifold the aspects which this

relation took on, shading from a clinging attachment to a passionate

conflict, the office of "philosophy" remained always that which

1 Cf., for example, Augustine, Solu. I. 7 ; Conf. V. 7; Scotus Erigena, DeDiv. Prcedest. I. 1 (Migne, 358) ; Anselm Proslog., ch. 1. (Migne, I. 227) ;

Abelard, Introd. in Theol. II. 3 ; Raymundus Lullus, De Quinque Sap. 8.2 Thomas Aquinas, Summa Theol. I. 32, 1 ; Contr. Gent. I. 8 f., II. 1 ff.

;

Duns Scotus, Op. Ox. I. 3, qu. 4;Durand de Pourc,ain, In Sent. Prol., qu. 8 ;

Raymundus of Sabunde, Theol. Natur. Prooem.8 Laur. Valla, Dialect. Disp. III. 9 ; B. Telesio, De Xat. Per. Prooem. ; Fr.

Bacon, De Auqm, III. 1 (Works, Spedding, I. 539 = 111. 336); Taurellus,Philos. Triumph. I. 1 ; Paracelsus, Parac/r. (ed. Huser) II. 23 f. ; G. Bruno.Delia Causa, etc., IV. 107 (Lagarde, I. 272) ; Hobbes, De Corpor. I. (Works,Molesworth. I. 2 and (! f.).

4 Characteristic definitions, on the one hand, in Gottsched, Erste Gründe der

gesammten Weltweisheit (Leips. 1756), pp. 97 ff. ; on the other hand, in thearticle Philosophie, in the Ency tiopedie (Vol. XXV. pp. 632 ff.).

Page 24: Windelband, History of Philosophy, Vol. 1

4 Introduction.

antiquity had assigned to it, to supply from scientific msight a

foundation for a theory of the world and of human life, where relig-

ion was no longer able to meet this need, or at least to meet it alone.

In the conviction that it was equal to this task, the philosophy of

the eighteenth century, like that of the Greeks, considered it its

right and duty to enlighten men with regard to the nature of things,

and from this position of insight to rule the life of the individual

and of society.

In this position of self-security philosophy was shaken by Kant,

who demonstrated the impossibility of a philosophical (i.e. meta-

physical) knowledge of the world beside of or above the individual

sciences, and thereby restricted once more the conception and the

task of philosophy ; for after this quitclaim the realm of philosophy,

as a particular science, was narrowed to just that critical consideration

by Reason of itself, from which Kant had won his decisive insight, and

which needed only to be extended systematically to activities other

than that of knowing. With this function could be united what

Kant 1 called the universal or cosmical conception of philosophy,—its vocation in the practical direction of life.

It is, to be sure, far from true that this new and apparently final

conception of philosophy gained universal acceptance at once. It is

rather the case that the great variety of philosophical movements of

the nineteenth century has left no earlier form of philosophy unre-

peated, and that a luxuriant development of the "metaphysical

need " 2 even brought back, for a time, the inclination to swallow up

all human knowledge in philosophy, and complete this again as an

all-embracing science.

2. In view of these mutations through which the meaning of the

word " philosophy " has passed in the course of time, it seems im-

practicable to pretend to gain a general conception of philosophy fromhistorical comparison. None of those brought forward for this

purpose 3 apply to all those structures of mental activity which

lay claim to the name. Even the subordination of philosophy under

the more general conception " science " is questionable in the case

of those types of teaching which place a one-sided emphasis on the

1 Critique of Pure Beason, A. 839; B. 866.2 Schopenhauer, World as Will and Idea, Vol. II. ch. 17.3 Instead of criticising particular conceptions it is sufficient here to point to

the widely diverging formulas in which the attempt has been made to performthis impossible task: cf., for example, only the introductions to works such as

those of Erdmann, Ueberweg, Kuno Fischer, Zeller, etc. All these conceptions

thus determined apply only in so far as the history of philosophy has yielded

the result which they express, but they do not apply with reference to the inten-

tions expressed by the philosophers themselves.

Page 25: Windelband, History of Philosophy, Vol. 1

§ 1.] Name and Conception of Philosophy. 5

practical significance of their doctrine :

lstill less can we define

the subject-matter and form of philosophy considered as a special

science, in a way that shall hold good for all cases. For even aside

from the primitive or the revived standpoint for which philosophy

is a universal science,2 the attempts to limit it are extremely vari-

ous. The problems of natural science form at first almost the sole

objects of interest for philosophy, then for a long period are in-

cluded in its scope, and do not separate from it until modern times.

History, on the other hand, has remained an object of indifference to

most philosophical systems, and has emerged as an object of philo-

sophical investigation relatively late and in isolated cases. Meta-

physical doctrines, again, in which the centre of philosophy is

usually sought, we see either pushed one side at important turning-

points in history or declared to be entirely impossible 3; and if at

times the ability of philosophy to determine the life of the indi-

vidual or of society is emphasised, a proud standpoint of pure theory

has renounced such a menial occupation.4

From still another side it has been claimed that philosophy treats

the same subjects as the other sciences, but in another sense and by

another method ; but neither has this specific characteristic of form

historical universality. That there is no such acknowledged his-

torical method would of course be no objection if only the endeavour

after such a method were a constant characteristic of all philoso-

phies. This is, however, so far from being the case that in fact

many philosophers imprint on their science the method of other

disciplines, e.g. of mathematics or of investigation of nature,5 while

others will have nothing at all to do with a methodical treatment of

their problems, and regard the philosophic activity as analogous to

the creations of genius in art.

3. From these circumstances is explained also the fact that there

is no fixed relation of philosophy to the other sciences, which is capa-

ble of a definition valid for all history. Where philosophy presents

itself as the universal science, the other sciences appear only as its

more or less distinctly separated parts.6 Where, on the contrary,

philosophy is assigned the task of grasping the results of the par-

1 So in the case of the majority of the philosophers of later antiquity.2 As for Chr. Wolf ; cf. his Logica, §§ 29 ff.

3 This is especially the case where philosophy is regarded solely as "scienceof cognition." Cf., e.g., W. Hamilton in his notes to Reid's works, II. 808.

Among the French at the close of the eighteenth and the beginning of this cen-

tury, philosophy = analyse de Ventendement humain.4 E.g. with Plotinus.5 So Descartes and Bacon.6 So, for example, in the Hegelian system.

Page 26: Windelband, History of Philosophy, Vol. 1

6 Introduction.

ticular sciences in their general significance, and harmonising theminto a comprehensive knowledge of the world, we have as the result

peculiarly complex relations : in the first place, a dependence of

philosophy upon the existing condition of insight reached in the par-

ticular disciplines— a dependence which expresses itself principally

in the furtherance of philosophy by the prominent advances made

by individual sciences ;x in the next place, an influence in the

opposite direction, when philosophy takes part in the work of the

particular sciences. This action is felt as help or as hindrance,

according as the philosophical treatment of the questions embraced

under the particular disciplines sometimes contributes valuable

factors for their solution, by means of its wider range of vision and

its tendency toward unity,2 but at other times presents itself only

as a duplication which, if it leads to like results, appears useless, or

if it wishes to furnish other results, dangerous. 3

From what has been said it is evident farther, that the relations

ofphilosophy to the other activities of civilisation are no less close than

its relation to the individual sciences. For the conceptions arising

from the religious and ethical and artistic life, from the life of the

state and of society, force their way everywhere, side by side with

the results won from scientific investigation, into the idea of the

universe which the philosophy of metaphysical tendencies aims to

frame ; and the reason's valuations ( Werthbestimmungen) and stand-

ards of judgment demand their place in that idea the more vigor-

ously, just in proportion as it is to become the basis for the practical

significance of philosophy. In this way humanity's convictions and

ideals find their expression in philosophy side by side with its

intellectual insights ; and if these convictions and ideals are regarded,

erroneously often, as gaining thereby the form of scientific intelli-

gence, they may receive under certain circumstances valuable clari-

fication and modification by this means. Thus this relation also of

philosophy to general culture is not only that of receiving, but also

that of giving.

It is not without interest to consider also the mutations in external positionand social relations which philosophy has experienced. It may be assumed that

science was from the first, with perhaps a few exceptions (Socrates), pursued in

Greece in closed schools. 4 The fact that these, even at a later time, had the form

1 As the influence of astronomy upon the beginnings of Greek, or that of

mechanics upon those of modern, philosophy.2 The Protestant theology of the nineteenth century stands in this relation

to German philosophy.3 Cf. the opposition of natural science to Schelling's philosophy of nature.4 H. Diels, Ueber die ältesten Philosophenschulen der Griechen in Philos.

Aufsätze zum Jubiläum E. Zeller's, Leips. 1887, pp. 241 ff.

Page 27: Windelband, History of Philosophy, Vol. 1

§ l.J Name and Conception of Philosophy. 7

of societies with religious laws l would not in itself alone, in view of the religious

character of all Greek judicial institutions, prove a religious origin of theseschools, but the circumstance that Greek science worked out its contents directly

from religious ideas, and that certain connections with religious cults presentthemselves unmistakably in a number of directions,'2 makes it not improbablethat the scientific societies sprang originally from religious unions (the Mys-teries) and continued in a certain connection with them. But when the scien-tific life had developed to complete independence, these connections fell awayand purely scientific schools were founded as free unions of men who, under theguidance of persons of importance, shared with each other the work of research,

exposition, defence, and polemic, 3 and at the same time had an ethical bond in

a common ideal of the conduct of life.

With the advent of the larger relations of life in the Hellenistic and Romanperiod, these unions naturally became loosened, and we frequently meet writers,

especially among the Romans, who are active in the field of philosophy in apurely individual way, neither members of a school nor professional teachers.

Such were Cicero, Seneca, and Marcus Aurelius. Not until the latest period of

antiquity were the ties of the schools drawn more closely again, as in Neo-Pythagoreanism and Neo-Platonism.Among the Romanic and Germanic peoples the course of events has been not

unlike that in the ancient world. The science of the Middle Ages also appearsin the train of the Church civilisation ; it has its seats in the cloister-schools, andis stimulated toward independent development primarily by questions of religious

interest. In it, too, the oppositions of various religious orders, such as the Do-minicans and Franciscans, assert themselves for a time, and even the freer

scientific associations out of which the universities gradually developed, hadoriginally a religious background and an ecclesiastical stamp.* Hence there

was always but a slight degree of independence with reference to Church doc-trine in this corporate philosophy of the universities, and this held true on into

the eighteenth century for the Protestant universities also, in the foundationand development of which ecclesiastical and religious interests had a foremostplace.

On the other hand, it is characteristic of the "world-wisdom" or secular

philosophy which was gaining its independence at the beginning of the modernperiod, that those who bring and support it are not at all men of the schools,

but men of the world and of life. An escaped monk, a state-chancellor, acobbler, a nobleman, a proscribed Jew, a learned diplomat, independent men of

letters and journalists, — these are the founders of modern philosophy, and in

accord with this, their work takes for its outer form not the text-book or the

deposit of academical disputations, but the free literary production, the essay.

Not until the second half of the eighteenth century did philosophy again

become corporate, and domesticated in the universities. This took place first

in Germany, where the most favourable conditions were afforded by the rising

independence of the universities, and where a fruitful interchange betweenteachers and students of the university was beneficial to philosophy also. 5

1 v. Wilamowitz-Möllendorf, Antigonns von Karystos (Philol. Stud. IV.Berlin, 1881, pp. 263 ft'.).

8 The Pythagoreans, as is well known, offer a pre-eminent example of this;

but sympathies with the Apollo cultus are plain enough in the Platonic Academyalso. Pfleiderer has lately sought to bring the apparently isolated Heraclitus

into connection with the Mysteries (E. Pfleiderer, Jleraklit von Ephesus.Berlin, 1886).

8 Cf. H. Usener, Ueber die Organisation der wisserischa/tlichen Arbeit imAlterthum (Preuss. Jahrb., Jahrg. LIIL, 1884, pp. 1 ff.)t and E. Heitz, Die Philo-

sophenschulen Athens (Deutsche Revue, 1884, pp. 326 if.).

4 Cf . G. Kaufmann, Geschichte der deutschen Universitäten I. pp. 98 ff. (Stuttg.

1888).6 Schelling has erected the finest monument to the ideal conception of science

in the activity of German universities, in his Vorlesungen über die Methode des

akademischen Studiums (2. and 3. Vorlesung. Ges. VVerke, I. Abth., Vol. 5.

pp. 223 ff.).

Page 28: Windelband, History of Philosophy, Vol. 1

8 Introduction.

From Germany this spread to Scotland, England, France, and Italy, and in gen-eral it may be said that in the nineteenth century the seat of philosophy is essen-

tially to be sought in the universities. 1

In conclusion, the share of the various peoples in the development of philoso-

phy deserves a brief mention. As with all developments of European culture,

so with philosophy, — the Greeks created it, and the primitive structure of

philosophy due to their creative activity is still to-day an essential basis of thescience. What was added in antiquity by the mixed peoples of Hellenism andby the Romans does not, in general, amount to more than a special form andpractical adaptation of the Greek philosophy. Only in the religious turn whichthis last movement took (cf. below, Part II. ch. 2) do we hnd something essen-

tially new which sprang from the harmonising of national differences in theRoman Empire. The scientific culture of the Middle Ages was also international,

as is implied in the universal employment of the Latin language. It is withmodern philosophy that the special characters of particular nations first presentthemselves as of decisive influence. While the traditions of mediseval scholas-

ticism maintain themselves most vigorously and independently in Spain andPortugal, the Italians, Germans, English, and French supply the first movementsof the new science which reached its highest point in the classical period of

German philosophy. Compared with these four nations, the rest stand almostentirely in a receptive attitude ; a certain independence is noticeable, if any-where, in more recent time among the Swedes.

§ 2. The History of Philosophy.

The more varied the character assumed by the problems and con-

tent of philosophy in the course of time, the more the question

rises, what meaning there can be in uniting in historical investiga-

tion and exposition products of thought which are not only so

manifold, but also so different in kind, and between which there

seems to be ultimately nothing in common but the name.

For the anecdotal interest in this checkered diversity of vari-

ous opinions on various things, which was perhaps formerly the

chief motive of a " History of Philosophy," stimulated too by the

remarkable and strange nature of many of these views, cannot

possibly serve as the permanent centre of a genuine scientific disci-

pline.

1. At all events, however, it is clear that the case stands other-

wise with the history of philosophy than with that of any other

science. For with all these the field of research remains fixed, on

the whole at least, however many the variations to which its extent,

its separation from a still more general field, and its limitation with

reference to neighbouring fields, may be subject in the course of his-

tory. In such a case there is no difficulty in tracing the develop-

ment of knowledge over a field which can be determined in this

way, and in eventually making just those variations intelligible as

the natural consequences of this development of insight.

1 The best evidence for this statement is afforded by just the passionateattacks which Schopenhauer directed against the relation between philosophyand the universities.

Page 29: Windelband, History of Philosophy, Vol. 1

§ 2.] The History of Philosophy. 9

Quite otherwise, however, in the case of philosophy, which has

no such subject-matter common to all its periods, and whose " his-

tory," therefore, sets forth no constant advance or gradual approxi-

mation to a knowledge of the subject in question. Rather, it has

always been emphasised that while in other sciences, a quiet build-

ing up of knowledge is the rule, as soon as they have once gained

a sure methodical footing after their rhapsodical beginnings, — a

rule which is interrupted only from time to time by a sudden newbeginning, — in philosophy the reverse is true. There it is the

exception that successors gratefully develop what has been already

achieved, and each of the great systems of philosophy begins to

solve its newly formulated problem ab ovo, as if the other systems

had scarcely existed.

2. If in spite of all of this we are still to be able to speak of a " his-

tory of philosophy," the unity of connection, which we find neither

in the objects with which philosophers busy themselves, nor in the

problems they have set themselves, can be found only in the commonivork which they have accomplished in spite of all the variety in their

subject-matter and in the purposes with which they have worked.

But this common product, which constitutes the meaning of the

history of philosophy, rests on just the changing relations which

the work of philosophers has sustained in the course of history, not

only to the maturest results of science in general and of the special

sciences in particular, but also to the other activities of European

civilisation. For wras it that philosophy had in view the project of

a general scientific knowledge of the universe, which she would win

either in the role of universal science, or as a generalising compre-

hension of the results of the special sciences, or was it that she

sought a view of life which should give a complete expression to

the highest values of will and feeling, or was it finally that with a

clearly defined limitation of her field she made reason's self-knowl-

edge her goal,— the result always was that she was labouring to

bring to conscious expression the necessary forms and principles in

which the human reason manifests its activity, and to transfer these

from their original form of perceptions, feelings, and impulses, into

that of conceptions. In some direction and in some fashion every

philosophy has striven to reach, over a more or less extensive field,

a formulation in conception of the material immediately given in

the world and in life ; and so, as these efforts have passed into his-

tory, the constitution of the mental and spiritual life has been

step by step disclosed. The History of Philosophy is the process in

lohich European humanity has embodied in scientific conceptions its

Views of the world and its judgments of life.

Page 30: Windelband, History of Philosophy, Vol. 1

10 Introduction.

It is this common fruit of all the intellectual creations which

present themselves as "philosophies," which alone gives to the

history of philosophy as a genuine science its content, its problem,

and its justification. This, too, is the reason why a knowledge of

the history of philosophy is a necessary requirement, not only for

all scholarly education, but for all culture whatever ; for it teaches

how the conceptions and forms have been coined, in which we all,

in every-day life as well as in the particular sciences, think and

judge the world of our experience.

The beginnings of the history of philosophy are to be sought in the historical

compositions (for the most part lost) of the great schools of antiquity, especially

the Peripatetic School. As we may see in the examples given by Aristotle, 1

these works had the critical purpose of preparing for the development of their

own views by a dialectical examination of views previously brought forward.

Such collections of historical material were planned for the various fields of

science, and doxographies 2 in philosophy arose in this way side by side withhistories of particular disciplines, such as mathematics, astronomy, physics, etc.

As inclination and power for independent philosophic thought later declined,

this literature degenerated into a learned scrap-book work, in which were mingledanecdotes from the lives of the philosophers, individual epigrammatic sayings,

and sketches of their doctrines.

Those expositions belonging to the modern period which were based uponthe remains of ancient tradition had this same character of collections of curiosi-

ties. Such were Stanley"1

s'A reproduction of Diogenes Laertius, and Brucker'sworks. 4 Only with time do we find critical discernment in use of the sources

(Buhle, 5 Fl'dleborn 6) , a more unprejudiced apprehension of the historical

significance of individual doctrines (Tiedemann," Degerando 8), and systematic

criticism of these upon the basis of the new standpoint (Tennemann,9 Fries, 10

and Schleiermacher 11) .

It was, however, through Hegel 12 that the history of philosophy was first

made an independent science, for he discovered the essential point that the

1 E.g. in the beginning of the Metaphysics.2 More in detail on these below.3 Th. Stanley, The History of Philosophy. Lond. 1685.4 J. J. Brucker, Historia Critica Philosophioe. 5 vols. Leips. 1742 ff. Insti-

tutiones Historian Philosophies. Leips. 1747.5 J. G. Buhle, Lehrbuch der Geschichte der Philosophie. 8 vols. Göttingen,

1796 ff.

6 G. G. Fülleborn, Beiträge zur Geschichte der Philosophie. 12 Studien.

Züllichau, 1791 ff.

7 D. Tiedemann, Geist der Specidativen Philosophie. 7 vols. Marburg,1791 ff.

8 De Gerando, Histoire Comparee des Systemes de Philosophie. 2d ed. in

4 vols. Paris, 1822 f.

9 W. G. Tennemann, Geschichte der Philosophie. 11 vols. Leips. 1798 ff.

Grundriss der Geschichte der Philosophie für den akademischen Unterricht.

Leips. 1812. [Eng. trans. 1833 and 1852.]lu J. Fr. Fries, Geschichte der Philosophie. 2 vols. Halle, 1837 ff.

11 Fr. Schleiermacher, Geschichte der Philosophie, from his literary remainsin the Coll. Works. III. Abth., 4 Bd., 1 Th. Berlin, 1839.

12 Cf. the introductions of the Phänomenologie des Geistes, of the lectures onthe Philosoph)/ of History, and those on the History of Philosophy. Ges. Werke,Bd. II. pp. 62 k.{ IX. pp. 11 ff. ; XIII. pp. 11-134. In Hegel's works the Geschichte

der Philosophic edited from his lectures by Michelet, occupies Vols. XIII.-XV.Berlin, 1833-36. [Lectures on the History of Philosophy, by G. W. Hegel.

Trans, by E. S. Haldane in 3 vols. Vol. I. Lond. 1892.J On his standpoint

Page 31: Windelband, History of Philosophy, Vol. 1

§2.] The History of Philosophy. 11

history of philosophy can set forth neither a motley collection of opinions ofvarious learned gentleman "de omnibus rebus et de quibusdam aliis," nor aconstantly widening and perfecting elaboration of the same subject-matter, butrather only the limited process in which the "categories" of reason have suc-cessively attained distinct consciousness and reached the form of conceptions.

This valuable insight was, however, obscured and injured in the case of Hegelby an additional asumption, since he was convinced that the chronological orderin which the above "categories" have presented themselves in the historicalsystems of philosophy must necessarily correspond with the logical and syste-matic order in which these same categories should appear as "elements oftruth " in the logical construction of the final system of philosophy {i.e. inHegel's view, his own). The fundamental thought, right in itself, thus led tothe mistake of a construction of the history of philosophy under the control of aphilosophical system, and so to a frequent violation of historical fact. Thiserror, which the development of a scientific history of philosophy in the nine-teenth century has set aside in favour of historical accuracy and exactness, arosefrom the wrong idea (though an idea in logical consistence with the principles ofHegel's philosophy) that the historical progress of philosophical thought is duesolely, or at least essentially, to an ideal necessity with which one "category"pushes forward another in the dialectical movement. In truth, the picture ofthe historical movement of philosophy is quite a different one. It depenas notsolely upon the thinking of "humanity " or even of the " Weltgeist,'''' but just

as truly upon the reflections, the needs of mind and heart, the presaging thoughtand sudden flashes of insight, of philosophising individuals.

3. The history of philosophy, considered as such a sum-total, in

which the fundamental conceptions of man's views of the world and

judgments of life have been embodied, is the product of a great

variety of single movements of thought. And as the actual motives

of these movements, various factors are to be distinguished, both in

the setting of the problems and in the attempts at their logical

solution.

The logical, pragmatic factor is no doubt sufficiently important.

For the problems of philosophy are in the main given, and this is

shown by the fact that they are constantly recurring in the histor-

ical movement of thought as the "primeval enigma of existence,"

and are ever anew demanding imperiously the solution which has

never completely succeeded. They are given, however, by the

inadequacy and internal contradictions of the material which con-

sciousness presents for philosophical consideration. 1 But just for

stand G. O. Marbach, Lehrbuch der Geschichte Philosophie (2. Abth. Leips.

1838 ff.), C. Hermann, Geschichte der Philosophie in pragmatischer Behandlung(Leips. 1867), and in part also the survey of the entire history of philosophywhich J. Braniss has published as the first (only) volume of a Geschichte der

Philosophie seit Kant (Breslau, 1842). In France this line is represented by V.Cousin, Introduction a VHistoire de la Philosophie (Paris, 1828 ; 7th ed. 1872)

;

Histoire Generale de la Philosophie (12th ed., Paris, 1884).1 More precisely, this inadequacy, which cannot here be more exactly devel-

oped, and which can be fully brought out only in a system of epistemology,

consists in the circumstance that that which is given in experience never meetscompletely the conceptional demands which, in elaborating the same accordingto the inner nature of the reason, we set up, at first naively and immediately,and later with reflective consciousness. This aiitinnmism (or failure to meetthe laws of thought) can be escaped by ordinary life, or even by experiential

Page 32: Windelband, History of Philosophy, Vol. 1

~\2 Introduction.

this reason this material contains the real presuppositions and the

logical constraining forces for all rational reflection upon it, and

because from the nature of the case these are always asserting

themselves anew in the same way, it follows that not only the chief

problems in the history of philosophy, but also the chief lines along

which a solution is attempted, are repeated. Just this constancy

in all change, which, regarded from without, makes the impression

that philosophy is striving fruitlessly in ever-repeated circles for

a goal that is never attained, proves only this,— that the problems

of philosophy are tasks which the human mind cannot escape. 1

And so we understand how the same logical necessity in repeated

instances causes one doctrine to give birth to another. Hence prog-

ress in the history of philosophy is, during certain periods, to be

understood entirely pragmatically, i.e. through the internal necessity

of the thoughts and through the " logic of things."

The mistake of Hegel's mentioned above, consists, then, only in his wishing to

make of a factor which is effective within certain limits, the only, or at least

the principal, factor. It would be the opposite error to deny absolutely the«« reason in history," and to see in the successive doctrines of philosophy onlyconfused chance-thoughts of individuals. It is rather true that the total content

of the history of philosophy can be explained only through the fact that the

necessities existing in the nature of things assert themselves over and over in

the thinking of individuals, however accidental the special conditions of this

latter may be. On these relations rest the attempts made to classify all philo-

sophical doctrines under certain types, and to establish a sort of rhythmicalrepetition in their historical development. On this basis V. Cousin 2 broughtforward his theory of the four systems, Idealism, Sensualism, Scepticism, Mys-ticism ; so too August Comte 3 his of the three stages, the theological, the meta-physical, and the positive. An interesting and in many ways instructive

grouping of philosophical doctrines about the particular main problems is

afforded by A. Itenouvier in his Esquisse d'une Classification Systematique

des Doctrines Philosophiques (2 vols., Paris, 1885 f.). A school-book whicharranges the philosophical doctrines according to problems and schools has beenissued by Paul Janet and Seailles ; Histoire de la Philosophie ; les problemes et

les ecoles (Paris, 1887).

4. But the pragmatic thread very often breaks off in the history

of philosophy. The historical order in particular, in which prob-

lems have presented themselves, shows almost a complete absence

science, by working with auxiliary conceptions, which indeed remain problem-atical in themselves, but which, within certain bounds, suffice for an elaboration

of the material of experience that meets our practical needs. But it is just in

these auxiliary conceptions that the problems of philosophy inhere.1 In this way the results of Kant's investigations on " The Antinomy of Pure

Reason " {Critique of Pure Reason, Transcendental Dialectic, second sec.) mightbe historically and systematically extended ; cf . W. Windelband, Geschichte der

neueren Philosophie, II. 95 f.

2 Cf. Note 12, p. 10.8 A. Comte, Cours de Philosophie Positive I. 9, with which Vols. V. and VI.

are to be compared as the carrying out of the scheme. Similar thoughts are

also found in D'Alembert's Discours Preliminaire in the Encyclopedie.

Page 33: Windelband, History of Philosophy, Vol. 1

§ 2.] The History of Philosophy. 1

3

of such an immanent logical necessity. Here, on the contrary,

another factor asserts itself which may best be designated as the

factor contributed by the history of civilisation. For philosophy

receives both its problems and the materials for their solution fromthe ideas of the general consciousness of the time, and from the

needs of society. The great conquests and the newly emergingquestions of the special sciences, the movements of the religious

consciousness, the intuitions of art, the revolutions in social andpolitical life, — all these give philosophy new impulses at irregular

intervals, and condition the directions of the interest which forces,

now these, now those, problems into the foreground, and crowds

others for the time being aside ; and no less do they condition also

the changes which questions and answers experience in course of

time. Where this dependence shows itself with especial clearness,

we have under certain circumstances a philosophical system appear-

ing, that represents exactly the knowledge which a definite age has

of itself ; or we may have the oppositions in the general culture of

the age finding their expression in the strife of philosophical sys-

tems. And so besides the constant dependence upon the essential

character of the subject-matter— the pragmatic factor— there pre-

vails also a necessity growing out of the history of civilisation, or

current state of culture, which warrants a historical right of exist-

ence to structures of thought in themselves untenable.

This relation also was first brought to notice in a greater degree than before

by Hegel, although the "relative truth" which he ascribes to the particular

systems has with him at the same time a systematic meaning, owing to his

dialectical fundamental thought. On the other hand, the element due to the

history of civilisation has been best formulated among his successors by KunoFischer, 1 who has also availed himself of it in most brilliant manner in his expo-

sition of the subject. He regards philosophy in its historical unfolding as the

progressive self-knowledge of the human mind, and makes its developmentappear as constantly conditioned by the development of the object which in it

is attaining self-knowledge. Although this applies to a number of the mostimportant systems, it is yet but one of the factors involved.

The influences from the history of civilisation which condition the statement

and solution of philosophic problems, afford an explanation in most cases of anextremely interesting phenomenon which is of great importance for understand-ing the historical development ; viz. the complication or interweaving of prob-

lems. For when interest is directed chiefly on certain lines of thought, it is

inevitable, according to psychological laws, that associations will be formedbetween different bodies of thought, — associations which are not based on the

subject-matter, — and so, that questions which in themselves have nothing to dowith each other become blended and made to depend upon each other in their

solution. An extremely important and very often recurring example of this is

the intermingling of ethical and aesthetic interests in the treatment of theoretical

problems. The well-known fact of daily life that men's views are determinedby their wishes, hopes, fears, and inclinations, that their theoretical are condi-

1 Kuno Fischer, Geschichte der neueren Philosophie, I. 1, Einleitung I.-V.

^trans. by J. P. Gordy, Descartes and his School, N.Y. 1887].

Page 34: Windelband, History of Philosophy, Vol. 1

14 Introduction.

tioned by their ethical and aesthetic judgments ( Urtheile durch ihre Beurthei-lungeii), — this fact is repeated on a larger scale in their views of the universe,

and has even been able to rise so high in philosophy that what had been pre-

viously involuntarily practised, was proclaimed (by Kant) an epistemological

postulate.

5. Meanwhile the historical process we are tracing owes all its

variety and multiplicity of forms to the circumstance that the de-

velopment of ideas and the formulation of general beliefs into

abstract conceptions are accomplished only through the thinking

of individual personalities, who, though rooted ever so deeply with

their thought in the logical connection and prevalent ideas of a

historical period, always add a particular element by their ownindividuality and conduct of life. This individual factor in the

development of the history of philosophy deserves so great atten-

tion for the reason that those who have borne the leading part in

the movement have shown themselves to be marked, independent

personalities, whose peculiar nature has been a determining in-

fluence, not merely for the selection and combination of problems,

but also for working out the conceptions to furnish solutions, both

in their own doctrines and in those of their successors. That history

is the kingdom of individualities, of details which are not to be

repeated and which have value in themselves, is shown also in the

history of philosophy : here, too, great personalities have exercised

far-reaching and not exclusively beneficial influences.

It is clear that the above-mentioned complication of problems is broughtabout by the subjective relations in which individual philosophers stand, in amuch greater degree than by the occasions presented in the general conscious-ness of a time, of a people, etc. There is no philosophical system that is free

from this influence of the personality of its founder. Hence all philosophicalsystems are creations of individuality, presenting in this respect a certain re-

semblance with works of art, and as such are to be understood from the point of

view of the personality of their founder. The elements of every philosopher'sWeltanschauung grow out of the problems of reality which are ever the same,and out of the reason as it is directed to their solution, but besides this out of

the views and ideals of his people and his time ; the form and arrangement,however, the connection and valuation which they find in the system, are condi-tioned by his birth and education, his activity and lot in life, his character andhis experience. Here, accordingly, the universality which belongs to the othertwo factors is often wanting. In the case of these purely individual creations,aesthetic charm must take the place of the worth of abiding knowledge, and theimpressiveness of many phenomena of the history of philosophy rests, in fact,

only upon the magic of their "poetry of ideas" (Begriffsdichtung).In addition, then, to the complication of problems and to the ideas determined

by fancy and feeling, which are already enough to lead the general conscious-ness astray, there are in the case of individuals similar, but purely personal,processes to lend to the formation and solution of problems still more the char-acter of artificiality. We cannot fail to recognise that philosophers have oftengone about struggling with questions which have no basis in reality, so that all

thought expended upon them was in vain, and that, on the other hand, even in

connection with the solution of real problems, unfortunate attempts in the apriori construction of conceptions have slipped in, which have been hindrancesrather than helps toward the issue of the matter.

Page 35: Windelband, History of Philosophy, Vol. 1

§ 2.] The History of Philosophy. 15

The wonderful feature in the history of philosophy remains just this, that

out of such a multitude of individual and general complications there has yetbeen on the whole laid down that outline of universally valid conceptions for

viewing the world and judging life, which presents the scientific significance of

this development.

6. Investigation in the history of philosophy has accordingly the

following tasks to accomplish: (1) To establish with precision what

may be derived from the available sources as to the circumstances

in life, the mental development, and the doctrines of individual

philosophers; (2) from these facts to reconstruct the genetic pro-

cess in such a way that in the case of every philosopher we mayunderstand how his doctrines depend in part upon those of his

predecessors, in part upon the general ideas of his time, and in part

upon his own nature and the course of his education; (3) from

the consideration of the whole to estimate what value for the total

result of the history of philosophy belongs to the theories thus

established and explained as regards their origin.

With reference to the first two points, the history of philosophy

is a philologico-historical, with reference to the third element it is a

critico-philosophical science.

(a) To establish its facts the history of philosophy must proceed to a careful

and comprehensive examination of the sources. These sources, however, varygreatly at different times in their transparency and fulness.

The main sources for investigation in the history of philosophy are of course

the works of the philosophers themselves. For the modern period we standhere upon a relatively safe footing. Since the discovery of the art of printing,

literary tradition has become so well established and clear that it offers in gen-eral no difficulties of any kind. The writings which philosophers have pub-lished since the Renaissance are throughout accessible for the research of

to-day. The cases in which questions of genuineness, of the time of origina-

tion, etc., give rise to controversies are extremely seldom ; a philological criti-

cism has here but a narrow field for activity, and where it can enter (as is thr

case in part in reference to the different editions of Kant's works), it concernssolely subordinate, and in the last instance indifferent, points. Here, too, we are

tolerably sure of the completeness of the material ; that anything of weight is

lost, or still to be expected from later publication, is scarcely to be assumed ; ii

the sharpened philological attentiveness of the last decades has brought us newmaterial for Spinoza, Leibniz, Kant, Maine de Biran, the philosophical outcomehas been only vanishing in comparison with the value of what was already

known. At most it has concerned the question of supplementing our knowl-edge, and this must continue to be its province. The importance of occasional

expressions in letters has been specially felt here, for these are adapted to shedmore light on the individual factor in the historical development of philosophy.

With the sources of the Mediaeval Philosophy the case stands less favourably.

These have in part (a small part, to be sure) still only a manuscript existence.

V. Cousin and his school have rendered valuable service in publishing the

texts, and in general we may be convinced that for this period also we possess

material, which has indeed gaps, but is on the whole adequate for our purpose.

On the other hand, our knowledge of the Arabian and Jewish philosophy of the

Middle Ages, and so of the influence of those systems on the course of WesternThought, is still very problematical in details ; and this is perhaps the gap mostsorely felt in our investigation of the sources for the history of philosophy.

Much worse still is the situation as regards the direct sources for AncientPhilosophy. Of the original works, we have preserved, to be sure, the most

Page 36: Windelband, History of Philosophy, Vol. 1

1(5 Introduction.

important : the fundamental portion of the works of Plato and Aristotle, thougheven these are often doubtful in form. Besides these we have cnly the writings

of later time, such as those of Cicero, Seneca, Plutarch, the Church Fathers,

and the Neo-Platonists. By far the greater part of the philosophical writings

of antiquity is lost. In their stead we must content ourselves with the frag-

ments which the accident of an incidental mention in the writings of extant

authors has kept for us, here too often in a questionable form.1

If, nevertheless, success has been attained in gaining a view of the develop-

ment of the ancient philosophy, clearer than that of the mediaeval, presenting apicture whose accuracy extends even to details and is scientifically assured, this

is due not only to the unremitting pains of philologists and philosophers in

working through their material, but also to the circumstance that beside the

remains of the original works of the philosophers there are preserved also, as

secondary sources, remains of historical records made in antiquity. The best,

indeed, of these also is lost: namely, the historical works which arose from the

learned collection made by the Peripatetic and Stoic schools at the end of the

fourth and in the third century b.c. These works passed later through manyhands before they were preserved for us in the extant compilations prepared in

the Roman period, as in the Placita Philosophorum? going by the name of

Plutarch, in the writings of Sextus Empiricus, 3 in the Deipnosophistce of Athe-

meus, 4 in the treatise of Diogenes Laertius, irepi ßluv doyfidruv xal d-rrodey/jidTuv

t&v iv <f>i\o<ro<plq. evdoKi/j.TjcrdvTwv,5 in the collections of the Church Fathers, andin the notes of the Commentators of the latest period, such as Alexander Aphro-

disias, Themistius, and Simplicius. H. Diels has given an excellent and thor-

ough treatment of these secondary sources of ancient philosophy, DoxographiGrceci (Berlin, 1879).Where the condition of the sources is so doubtful as is the case over the

entire field of ancient philosophy, critical ascertainment of the facts must gohand in hand with examination of the pragmatic and genetic connection. Forwhere the transmission of the material is itself doubtful we can reach a decision

only by taking a view of the connection that shall accord with reason andpsychological experience. In these cases it becomes the task of the history of

philosophy as of all history, after establishing a base of operations in that whichis assured by the sources, to proceed to ascertain its position in those regions

with which tradition finds itself no longer directly and surely in touch. Thehistorical study of philosophy in the nineteenth century may boast that it hasfulfilled this task, to which it was stimulated by Schleiermacher, by the laboursof H. Eitter, — whose Geschichte der Philosophie (12 vols., Hamburg, 1829-53) is

now, to be sure, antiquated, — Brandis and Zeller for the ancient philosophy;

and of J. E. Erdmann and Kuno Fischer for the modern. Among the manycomplete expositions of the history of philosophy by far the most trustworthyin these respects is J. 3. Erdmann's Orundriss der Geschichte der Philosophie,

2 vols. (3d ed.), Berlin, 1878;[Erdmann's History of Philosophy, trans, ed. by

W. S. Hough, Lond. and N.Y., 1890].

An excellent bibliography of the entire history of philosophy, assembling the

literature in exhaustive completeness and good arrangement, is to be found in

Ueberweg's Grundriss der Geschichte der Philosophie, 4 vols., 8th ed., ed. byM. Heinze (Berlin, 1894-98). [Ueberweg's History of Philosophy, trans, fromthe 4th ed. by G. S. Morris (N. Y. 1871), contains additions, but of course does not

1 The collections of fragments of particular authors are mentioned under the

notices of the individual philosophers. It would be desirable if they were all as

excellent as Usener's Epicurea. Of the fragments of the Pre-Socratics W. F.

A. Mullach has published a careful collection, which, however, is no longer

adequate in the present condition of research (Fragmenta PhilosophorumGrcecorum).

2 Plut. Moralia, ed. Dübner, Paris, 1841 ; Diels, Box., pp. 272 ff.;[Plutarch's

Morals, Miscellanies, and Essays, ed. by Goodwin, Boston, 1870 ; trans, also in

the Bohn Lib.].3 Ed. Bekker, Berlin, 1847.* G. Kaibel, Leips. 1888-90.6 Ed. Cobet, Paris, 1850.

Page 37: Windelband, History of Philosophy, Vol. 1

§ 2.] The History of Phüosophj. 1

7

give the bibliography of recent works.] Under the general literature may alsobe mentioned, R. Eucken, Die Lebensanschauungen der grossen Denker (Leins.1890).

(&) Explanation of facts in the history of philosophy is either pragmatic (logi-

cal), or based on the history of civilisation, or psychological, corresponding to th«three factors which we have set forth above as determining the movement oJ

thought. Which of these three modes of explanation is to be applied in individ.ual cases depends solely upon the state of the facts with regard to the trans-mission of material. It is then incorrect to make either one the sole principleof treatment. The pragmatic method of explanation is dominant with thosewho see in the entire history of philosophy the preparation for a definite systemof philosophy ; so with Hegel and his disciples (see above, p. 10 f.); so from aHerbartian standpoint with Chr. A. Thilo, Kurze pragmatische Geschichte derPhilosophie (2 pts. ; Coethen, 1876-80). Kuno Fischer and W. Windelbandhave emphasised in their interpretation of modern philosophy, the importanceof considering the history of civilisation and the problems of the individualeciences.

The purely biographical treatment which deals only with successive person-alities is quite inadequate as a scientific exposition of the history of philosophy.This mode of treatment is represented in recent time by the treatise of G. H.Lewes, The History of Philosophy from Thales to the Present Day (2 vols.,

Lond. 1871), a book destitute of all historical apprehension, and at the sametime a party composition in the spirit of the Positivism of Comte. The worksof the French historians (Damiron, Ferraz) are inclined to take this form ofa separate essay-like treatment of individual philosophers, not losing from sight,

however, the course of development of the whole. 1

(c) The most difficult task is to establish the principles according to which thecritical philosophical estimate of the individual doctrines must be made up.The history of philosophy, like all history, is a critical science ; its duty is notonly to record and explain, but also to estimate what is to count as progressand fruit in the historical movement, when we have succeeded in knowing andunderstanding this. There is no history without this critical point of view, andthe evidence of a historian's maturity is that he is clearly conscious of this pointof view of criticism ; for where this is not the case he proceeds in the selectionof his material and in his characterisation of details only instinctively andwithout a clear standard.'-*

It is understood, of course, that the standard of critical judgment must not bea private theory of the historian, nor even his philosophic conviction ; at least

the employment of such a standard deprives the criticism exercised in accord-ance with it of the value of scientific universality. He who is given to thebelief that he possesses the sole philosophical truth, or who comes to this field

imbued with the customs of the special sciences in which, no doubt, a sure result

makes it a very simple 3 matter to estimate the attempts which have led to it,—such a one may well be tempted to stretch all forms that pass before him uponthe Procrustes-bed of his system ; but he who contemplates the work of thoughtin history, with an open historical vision, will be restrained by a respectfulreverence from reprimanding the heroes of philosophy for their ignorance of thewisdom of an epigone. 4

1 A. Weber, History of Philosophy, is to be recommended as a good text-book(5th French ed., Paris, 1891). [Eng. tr. by Thilly, N.Y. 1896.]

2 This applies in every domain of history, in the history of politics and of

literature, as well as in that of philosophy.3 As an example of this it may be noticed that the deserving author of an

excellent History of the Principles of Mechanics, Ed. Diihring, has developedin his Kritische Geschichte der Philosophie (."3d ed., Berlin. 1878) all the capriceof a one-sided judgment. The like is true of the confessional criticism passedby A. Stöckl, Lehrbuch der Geschichte der Philosophie (2 vols., 3d ed., Mainz,1889).

* It is impossible to protest enough against the youthful conceit with whichIt was for a time the fashion in Germany to look down with ridicule or insult

from the " achievements of the present " upon the great men of Greek and < rer-

Page 38: Windelband, History of Philosophy, Vol. 1

18 Introduction.

In contrast with this external method of pronouncing senterce, the scientific

history of philosophy must place itself upon the standpoint of immanent criti-

cism, the principles of which are two : formal logical consistency and intellectual

fndtfulness.Every philosopher grows into a certain set of ideas, and to these his thinking

remains bound, and is subjected in its development to psychological necessity.

Critical investigation has to settle how far it has been possible for him to bringthe different elements of his thinking into agreement with each other. Thecontradiction is almost never actually present in so direct a form that the samething is expressly maintained and also denied, but always in such a way thatvarious positions are put forward which, only by virtue of their logical conse-

quences, lead to direct contradiction and really irreconcilable results. The dis-

covery of these discrepancies is formal criticism ; it frequently coincides withpragmatic explanation, for this formal criticism has been performed in history

itself by the successors of the philosopher in question, and has thus determinedfor them their problems.Yet this point of view alone is not sufficient. As purely formal it applies

without exception to all attested views of a philosopher, but it gives no criterion

for decision on the question, in what the philosophical significance of a doctrine

really consists. For it is often the case that philosophy has done its work just

in conceptions which must by no means be regarded as in themselves perfect

or free from contradiction ; while a multitude of individual convictions, whichthere is no occasion to oppose, must remain unnoticed in a corner, so far as ourhistorical survey is concerned. In the history of philosophy great errors are

weightier than small truths.

For before all else the decisive question is : what has yielded a contribution to

the development of man's conception of the universe and estimate of life ? Inthe history of philosophy those structures of thought are the objects of studywhich have maintained themselves permanent and living as forms of apprehen-sion and norms of judgment, and in which the abiding inner structure of thehuman mind has thus come to clear recognition.

This is then the standard, according to which alone we can decide also whichamong the doctrines of the philosophers— concerning, as they often do, somany various things— are to be regarded as properly philosophical, and which,on the other hand, are to be excluded from the history of philosophy. Investi-

gation of the sources has of course the duty of gathering carefully and com-pletely all the doctrines of philosophers, and so of affording all the material for

explaining their genesis, whether from their logical content, or from the history

of civilisation, or from psychological grounds ; but the purpose of this laborious

work is yet only this, that the philosophically indifferent may be ultimately

recognised as such, and the ballast then thrown overboard.It is especially true that this point of view must essentially determine selec-

tion and presentation of material in a text-book, which is not to give the investi-

gation itself, but to gather up its results.

§ 3. Division of Philosophy and of its History.

It cannot be our purpose here to propose a systematic division of

philosophy, for this could in no case possess universal validity his-

torically. The differences which prevail in the course of the histori-

cal development, in determining the conception, the task, and the

subject-matter of philosophy, involve so necessarily and obviously a

change also in the divisions, that this needs no especial illustration.

The oldest philosophy knew no division at all. In later antiquity

man philosophy ; this was mainly the haughtiness of an ignorance which hadno suspicion that it was ultimately living only by the thoughts of those whom it

was abusing and despising.

Page 39: Windelband, History of Philosophy, Vol. 1

§ 3.] -Division of Philosophy and of its History. 19

a division of philosophy into logic, physics, and ethics was cur-

rent. In the Middle Ages, and still more in modern times, the

first two of these subjects were often comprised under the title,

theoretical philosophy, and set over against practical philosophy.

Since Kant a new threefold division into logical, ethical, and

ajsthetical philosophy is beginning to make its way, yet these

various divisions are too much dependent upon the actual course

of philosophy itself to make it worth our while to recount themhere in detail.

On the other hand, it does commend itself to preface the historical

exposition with at least a brief survey of the entire circuit of those

problems which have always formed the subject of philosophy, how-

ever varied the extent to which they have been studied or the value

that has been attached to them, — a survey, therefore, for which no

claim is made to validity from a systematic point of view, but which

is determined only by the purpose of preliminary orientation.

1. Theoretical problems. Such we call those which refer, in part to

our knowledge of the actual world, in part to an investigation of the

knowing process itself. In dealing with the former class, however,

the general questions which concern the actual taken as a whole are

distinguished from those which deal with single provinces of the

actual. The former, viz. the highest principles for explaining the

universe, and the general view of the universe based on these prin-

ciples, form the problem of metaphysics, called by Aristotle first, i.e.

fundamental, science, and designated by the name now usual, only on

account of the position which it had in the ancient collection of the

Aristotelian works — " after physics." On account of his monothe-

istic view of the world, Aristotle also called this branch of knowl-

edge theology. Later writers have also treated rational or natural

theology as a branch of metaphysics.

The special provinces of the actual are Nature and History. In

the former, external and internal nature are to be distinguished.

The problems presented to knowledge by external nature are called

cosmological, or, specially, problems of natural philosophy, or perhaps

•physical. The investigation of internal nature, i.e. of consciousness

and its states and activities, is the business of j>sychology. The phil-

osophical consideration of history remains within the borders of

theoretical philosophy only if it be limited to the investigation of

the laws that prevail in the historical life of peoples ; since, how-

ever, history is the realm of man's purposeful actions, the questions

of the philosophy of history, so far as this deals with the end of the

movement of history viewed as a whole, and with the fulfilment of

this end, fall under the head of practical problems.

Page 40: Windelband, History of Philosophy, Vol. 1

20 Introauacion.

Investigation directed upon knowledge itself is called logic (in

the general sense of the word), and also sometimes noetic. If weare occupied with the question how knowledge actually arises, this

psycho-genetic consideration falls in the province of psychology. If,

on the other hand, we set up norms or standards according to which

our ideas are estimated as regards their worth for truth, we call

these logical laws, and designate investigation directed upon them

as logic in the narrower sense. The application of these laws gives

rise to methodology, which develops the prescriptions for a systematic

ordering of scientific activity with reference to the various ends of

knowledge. The problems, finally, which arise from the questions

concerning the range and limit of man's knowing faculty and its

relation to the reality to be known, form the subject-matter of

epistemology or theory of"knowledge.

H. Siebeck, Geschichte der Psychologie, Vol. I., in two parts (Gotha, 1880-84),incomplete, extending into the scholastic period.

K. Prantl, Geschichte der Logik im Abendlande, 4 vols. (Leips. 1855-70)v

brought down only to the Renaissance.

Fr. Harms, Die Philosophie in ihrer Geschichte. I. "Psychologie"; II.

"Logik" (Berlin, 1877 and 1881).

[R. Adamson, The History of Psychology (in prep.).]

2. Practical problems are, in general, those which grow out of the

investigation of man's activity, so far as it is determined by ends.

Here, too, a psycho-genetic treatment is possible, which falls under

psychology. That discipline, on the other hand, which considers

man's action from the point of view of the ethical norm or stand-

ard, is ethics or moral philosophy. By morals {Moral) in the narrower

sense is usually understood the proposal and grounding of ethical

precepts. Since, however, all ethical action has reference to the

community, there are attached to morals or ethics, in the narrower

sense, the philosophy of society (for which the unfortunate name

sociology seems likely to become permanent), and the philosophy of

law or right. Further, in so far as the ideal of human society con-

stitutes the ultimate meaning of history, the philosophy of history

appears also in this connection, as already mentioned.

To practical problems, in the broadest sense of the word, belong

also those which relate to art and religion. To designate philosoph-

ical investigation of the nature of the beautiful and of art, the name

aesthetics has been introduced since the end of last century. If phi-

losophy takes the religious life for its object, not in the sense of

itself intending to give a science of the nature of the deity, but in

the sense of an investigation with regard to man's religious behaviour,

we call this discipline philosophy of religion.

Page 41: Windelband, History of Philosophy, Vol. 1

§ 3.] Division of Philosophy and of its History

.

2!

Fr. Schleiermacher, Grundlinien einer Kritik der bisherigen Sittenlehre (col-

lected works, III., Vol. I., Berlin, 1834). L. v. Henning, Die Principien derEthik in historischer Entwicklung (Berlin, 1825). Fr. v. Raumer, Die ge-schichtliche Entwicklung der Begriffe von Staat, Recht, und Politik (Leips., 3ded., 1861). E. Feuerlein, Die philos. Sittenlehre in ihren geschichtlichen Haupt-formen (2 vols., Tübingen, 1857-59). P. Janet, Histoire de la Philosophiemorale et politique (Paris, 1858). W. Whewell, History of Moral Science(Edinburg, 1863). H. Sidgwick, The Methods of Ethics, 4th ed. (Loud, andN.Y. 1890). [Outlines of the History of Ethics, by same author (Lond. andN.Y., 3d ed., 1892). J. Martineau, Types of Ethical Theory (2d ed., Oxfordand N.Y. 1886).] Th. Ziegler, Geschichte der Ethik, 2 vols, (the third not yetappeared; Strassburg, 1881-86). K. Köstlin, Geschichte der Ethik (only thebeginning, 1 vol., Tübingen, 1887). [J. Bonar, Philosophy and Economics in

their Historical Relations (Lond. and N.Y. 1893). D. G. Ritchie, The History

of Political Philosophy (in prep.).]

R. Zimmermann, Geschichte der Aesthetik (Vienna, 1858). M. Schasler,

Kritische Geschichte der Aesthetik (Berlin, 1871). [B. Bosanquet, The History

of ^Esthetics (Lond. and N.Y. 1892). W. Knight, The Philosophy of the Beau-tiful (an outline of the history, Edin. and N.Y. 1891). Gayley and Scott, AGuide to the Literature of ^Esthetics, Univ. of California, and Introd. to thtMethodsand Materials of Literary Criticism (Bost. 1899) have bibliographies.]

J. Berger, Geschichte der Religionsphilosophie (Berlin, 1800). [Pünjer,History of the Christian Philosophy of Religion (Vol. I., Edin. and N.Y. 1887).0. Ffleiderer, The Philosophy of Religion, trans, by Menzies (Lond. 1887). Mar-tineau, A Study of Religion (2 vols., 1888), and Seat of Authority in Religion(1890). J. Caird, Introd. to the Philos. of Religion (1880). E. Caird, Evolu-tion of Religion (2 vols., Lond. and N.Y. 1893).]

The division of the history of philosophy is usually connected with

that current for political history, so as to distinguish three great

periods,— Ancient, Mediaeval, and Modern Philosophy. Yet the

sections made in this way are not so favourable for the history of

philosophy as they perhaps are for political history. Other points

of division must be made, equally important as regards the nature

of the development ; and, on the other hand, the transition between

the Middle Ages and modern times demands a shifting of the point

of division on either side.

In consequence of this, the entire history of philosophy will here

be treated according to the following plan of division, in a manner

to be more exactly illustrated and justified in detail by the exposi

tion itself :—

(1) Tlie Philosophy of the Greeks: from the beginnings ol

scientific thought to the death of Aristotle,— from about 600 to

322 b.c.

(2) Hellenistic-Roman Philosophy : from the death of Aristotle

to the passing away of Neo-Platonisin,— from 322 b.c. to about

\500 A.D.

(3) Mediozval Philosophy : from Augustine to Nicolaus Cusanus,

— from the fifth to the fifteenth century.

(4) The Philosophy of the Renaissance : from the fifteenth to the

seventeenth century.

Page 42: Windelband, History of Philosophy, Vol. 1

22 Introduction.

(5) The Philosophy of the Enlightenment: from Locke to thedeath of Lessing,— 1689-1781.

(6) The German Philosophy : from Kant to Hegel and Herbart,—1781-1820.

(7) The Philosophy of the Nineteenth Century.

Page 43: Windelband, History of Philosophy, Vol. 1

PART I.

THE PHILOSOPHY OF THE GREEKS.

Chr. A. Brandis, Handbuch der Geschichte der griechisch-römischen Philosophie.

3 pts. in 6 vols. Berlin, 1835-66.

Same author, Geschichte der Entwickelimgen der griechischen Philosophie undihrer Nachwirkungen im römischen Reiche. 2 pts. Berlin, 1862-66.

Ed. Zeller, Die Philosophie der Griechen. 3 pts. in 5 vols. 1st vol. in 5th,

2 vol. in 4th, 3-5 vols, in 3d ed. Leips. 1879-93. [Trans., with the excep-

tion of the portion on the concluding religious period, as six works : Pre-

Socratic Philosophy (2 vols.), Socrates and the Socratic Schools, Plato and

the Older Academy, Aristotle and the Earlier Peripatetics (2 vols.), Stoics.

Epicureans, and Sceptics, History of Eclecticism, chiefly by S.F. Alleyne and

O. J. Reichel. Lond. and NY., Longmans.]

A. Schwegler, Geschichte der griechischen Philosophic Ed. by K. Köstlin. 3d

ed. Freiburg, 1882.

L. Strümpell, Die Geschichte der griechischen Philosophie. 2 pts. Leips.

1854-61.

W. Windelband, Geschichte der alten Philosophie. 2d ed. Munich, 1894.

[History of Ancient Philosophy, trans, by H. E. Cushman, N.Y., 1899. J

Ritter et Preller, Historia philosophies, grceco-romanoe (Grcecce). In 8th ed.

Edited by Wellman. Gotha, 1898. An excellent collection of the most

important sources.

[A. W. Benn, The Greek Philosophers. 2 vols. Lond., 1883. The Philoso-

phy of Greece. Lond. 1898.]

Th. Gomperz, Griechische Denker. Vienna, 1897. [Trans, by L. Magnus.

Greek Thinkers. Lond. and N.Y., 1900.]

If by science we understand that independent and self-conscious

work of intelligence which seeks knowledge methodically for its

own sake, then it is among the Greeks, and the Greeks of the sixth

century b.c., that we first find such a science,— aside from some

tendencies among the peoples of the Orient, those of China and

India 1 particularly, only recently disclosed. The great civilised

1 Even if it be conceded that the beginnings of moral philosophy among the

Chinese rise above moralising, and especially those of logic in India above inci-

dental reflections on the scientific formation of conceptions, — on which we shall

not here pronounce, — these remain so remote from the course of Europeanphilosophy, which forms a complete unity in itself, that a text-book has nooccasion to enter upon them. The literature is brought together in Ueber-weg, I. § 6.

23

Page 44: Windelband, History of Philosophy, Vol. 1

24 The Philosophy of the Greeks.

peoples of earlier antiquity were not, indeed, wanting either in an

abundance of information on single subjects, or in general views of

the universe ; but as the former was gained in connection with prac-

tical needs, and the latter grew out of mythical fancy, so they

remained under the control, partly of daily need, partly of religious

poetry ; and, as was natural in consequence of the peculiar restraint

of the Oriental mind, they lacked, for their fruitful and independent

development, the initiative activity of individuals.

Among the Greeks, also, similar relations existed until, at the time

mentioned, the mighty upward movement of the national life unfet-

tered the mental powers of this most gifted of all peoples. For this

result the democratic development of constitutions which in passion-

ate party struggle tended to bring out independence of individual

opinions and judgments, and to develop the significance of person-

ality, proved even more favourable than the refinement and spiritual-

isation of life which increasing wealth of trade brought with it.

The more the luxuriant development of individualism loosened the

old bonds of the common consciousness, of faith, and of morals, and

threatened the youthful civilisation of Greece with the danger of

anarchy, the more pressing did individual men, prominent by their

position in life, their insight, and their character, find the duty

of recovering in their own reflection the measure that was becoming

lost. This ethical reflection found its representatives in the lyric

and gnomic poets, especially, however, in the so-called seven ivise men. 1

It could not fail to occur, also, that a similar movement, in which

individual opinions asserted their independence, should trench upon

the religious life already so varied, in which the opposition between

the old mystery-cults and the aesthetic national mythology stimu-

lated the formation of so many special types. 2 Already in the cos-

mogonic poetry the poet had dared to portray the heaven of the

myths according to his own individual fancy ; the age of the seven

sages began to read its ethical ideals into the gods of the Homeric

poetry, and in the ethico-religious reformation attempted by Pythag-

oras,3 coming as it did in the outer form of a return to the old strict-

ness of life, the new content which life had gained came all the more

clearly to view.

1 The "seven sages," among whom Thales, Bias, Pittacus, and Solon are

usually named, while with regard to the rest tradition is not agreed, must not,

with the exception of Thales, be regarded as representatives of science. Diog.

Laert. I. 40 ; Plato, Protag. 343.2 Cf. E. Rohde (Psyche, 2d ed., 1897) for the influence of religious ideas.3 Pherecydes of Syrus is to be regarded as the most important of these cos-

mogonic poets ; he wrote in prose at the time of the first philosophies, but his

mode of thought is still mythical throughout, not scientific. Fragments of his

writings collected by Sturz (Leips. 1834).

Page 45: Windelband, History of Philosophy, Vol. 1

The Philosophy of the Greeks. 25

From such conditions of fermentation the science of the Greeks

to which they gave the name philosophy was born. The independ-

ent reflection of individuals, aided by the fluctuations of religious

fancy, extended itself from the questions of practical life to the

knowledge of Nature, and there first won that freedom from exter-

nal ends, that limitation of knowledge to itself, which constitutes

the essence of science.

All these processes, however, took place principally in the outly-

ing parts of Greek civilisation, in the colonies, which were in advance

of the so-called Mother-country in mental as in material develop-

ment. In Ionia, in Magna Graecia, in Thrace, stood the cradles of

science. It was only after Athens in the Persian wars had assumed

together with the political hegemony the mental as well, which she

was to keep so much longer than the former, that Attic soil, conse-

crated to all the muses, attracted science also. Its advent was at

the time of the Sophists ; it found its completion in the doctrine

and school of Aristotle.

It was in connection with the disinterested consideration of

Nature that reflection first rose to the scientific construction of

conceptions. The result of this was that Greek science devoted all

the freshness of youthful joy and knowledge primarily to the prob-

lems of Nature, and in this work stamped out fundamental concep-

tions, or Forms of thought, for apprehending the external world. In

order to turn the look of philosophy inward and make human action

the object of its study, there was first need, for one thing, of subse-

quent reflection upon what had, and what had not, been accomplished

by this study of Nature, and, for another thing, of the imperious

demands made by public life on science now so far matured as to be

a social factor. The effect of this change might for a time seem to

be to check the pure zeal for research which had marked the begin-

nings, but after positive results had been reached in the field of the

knowledge of man's inner nature this same zeal developed all the

more vigorously, and led to the construction of those great systems

with which purely Greek philosophy reached its consummation.

The philosophy of the Greeks divides, therefore, into three jJeriods :

a cosmological, which extends from about 600 to about 450 b.c. ;an

anthropological, which fills out about the second half of the fifth

century b.c. (450-400) ; and a systematic, which contains the

development of the three great systems of Greek science, those of

Democritus, Plato, and Aristotle (400-322).

The philosophy of the Greeks forms the most instructive part of the wholehistory of philosophy from a theoretical point of view, not only because the

fundamental conceptions created in it have become the permanent foundations

Page 46: Windelband, History of Philosophy, Vol. 1

26 The Philosophy of the Greeks.

for all further development of thought, and promise to remain such, but also

because in it the formal presuppositions contained in the postulates of the

thinking Reason itself, attained sharp formulation as set over against the mate-rial of knowledge, which, especially at the beginning, was still relatively small

in amount. In this the Greek philosophy has its typical value and its didactic

importance.These advantages appear already in the transparency and simplicity of the

entire development, which enable us to see the inquiring mind at first turnedoutward, then thrown back upon itself, and from this point of view returning

to a deeper apprehension of reality as a whole.

There is, therefore, scarcely any controversy with regard to this course of

the general development of Greek philosophy, though different expositions havelocated the divisions between the periods at different points. Whether Socrates

is made to begin a new period, or is placed together with the Sophists in the

period of Greek Enlightenment, depends ultimately only on whether the result

(negative or positive), or the object-matter of the philosophising, is regarded as.

of decisive importance. That, however, Democritus must in any case be sepa-rated from the " Pre-Socratics " and assigned to the great systematic period

of Greek Philosophy, has been proved by the Author in his survey of the

History of Ancient Philosophy, ch. V., and the objections which the innovationhas encountered have not sufficed to convince him of any mistake.

Page 47: Windelband, History of Philosophy, Vol. 1

CHAPTER I.

THE COSMOLOGICAL PERIOD.

S. A. Byk, Die vorsokratische Philosophie der Griechen in ihrer organischen

Gliederung. 2 Parts. Leips. 1875-77.

[J. Burnet, Early Greek Philosophy. Lond. 1892.]

The immediate background for the beginnings of Greek philoso-

phy was formed by the cosmogonic poetry, which aimed to present

in mythical garb the story of the prehistoric ages of the given

world, and so, in the form of narratives of the origination of the

universe, made use of prevailing ideas as to the constant mutations

of things. The more freely individual views developed in this pro-

cess, the more the time factor in the myth retreated in favour of the

emphasising of these abiding relations ; and the question finally

emerged : " What is then the original ground of things, which out-

lasts all temporal change, and how does it change itself into these

particular things, or change these things back into itself ? "

The solution of this question was first attempted in the sixth

century by the Milesian School of natural philosophy, of which

Thales, Anaximander, and Anaximenes are known to us as the

three chief representatives. Information of many kinds, Avhich had

long been publicly accumulating m the practical experience of the

sea-faring Ionians, stood at their disposal, as well as many true

observations, often of an acute sort. They kept in touch, also, no

doubt, with the experience of the Oriental peoples, especially the

Egyptians, with whom they stood in so close relation. 1 Knowledge

from these various sources was brought together with youthful zeal.

The chief interest fell upon physical questions, particularly upon

1 The influence of the Orient upon the beginnings of Greek philosophy hasbeen overestimated by Glabisch (Die Religion und die Philosophie in ihrer

loeltgeschichtlichen Enticicklung, Breslau, 1882) and Both (Geschichte unsere?abendländischen Philosophie, 2 Vols., Mannheim, 1858 ff.). In the case of

information upon particular fields such influence is certainly to be recognised;

on the other hand, the scientific conceptions are throughout independent worksof Greek thought.

27

Page 48: Windelband, History of Philosophy, Vol. 1

28 The Philosophy of the Greeks. [Part I.

the great elementary phenomena, to explain which many hypotheses

were thought out. Besides this, interest turned chiefly to geo-

graphical and astronomical problems, such as the form of the earth,

its relation to the sidereal heavens, the nature of the sun, moon,

and planets, and the manner and cause of their motion. On the

other hand, there are but feeble indications of a zeal for knowledge

applied to the organic world and man.

Such were the objects of experience studied by the first "philosophy." It

stood quite far removed from medical science, which, to be sure, was limited totechnical information and proficiency in the art, and was handed down as asecret doctrine, guarded in priest-like fashion in orders and schools, such asthose of Rhodes, Cyrene, Crotona, Cos, and Cnidus. Ancient medicine, whichaimed expressly to be an art and not a science (so Hippocrates), came into

contact with philosophy when this was an all-embracing science, only at a late

period and quite transiently. Cf. Häser, Lehrbuch der Geschichte der Medicin,I. (2d ed., Jena, 1875).

So also the beginnings of mathematics go along independently beside those of

ancient philosophy. The propositions ascribed to the Milesians make the im-pression of individual pieces of information picked up and put together, ratherthan of results of genuine research, and are quite out of relation with their

doctrines in natural science and philosophy. In the circles of the Pythagoreans,also, mathematical studies were at first evidently pursued for their own sake, tobe drawn all the more vigorously into the treatment of general problems. Cf.

G. Cantor, Geschichte der Mathematik, I. (Leips. 1880).

The efforts of the Milesians to determine the nature of the one

world-ground had already in the case of Anaximander led beyond

experience to the construction of a metaphysical conception to be

used for explanation, viz. the a-n-apov, and thereby drew science awayfrom the investigation of facts to the consideration of conceptions.

While Xenophanes, the founder of the Eleatic School, drew the con-

sequences which result for the religious consciousness from the

philosophical conception of the unity of the world, Heraclitus, in

hard struggle with ideas that were obscure and religiously coloured,

analysed destructively the presupposition of an abiding substance,

and allowed only a law of change to stand as ultimate content of

knowledge. All the more sharply, on the other hand, did the Eleatic

School, in its great representative, Parmenides, shape out the con-

ception of Being until it reached that regardless boldness of formu-

lation which, in the following generation of the School, was defended

by Zeno, and softened down in some measure only by Melissus.

Very soon, however, a series of efforts appeared, which brought

anew into the foreground the interest in explanatory natural science

that had been thrust aside by this development of the first meta-

physical antitheses. In behalf of this interest more comprehensive

efforts were made toward an enrichment of knowledge ; this time,

more than in the case of previous observations, questions and

hypotheses from the organic and physiological realms were kept in

Page 49: Windelband, History of Philosophy, Vol. 1

Chap. 1.) The Cosmological Period. 29

mind; and the attempt was made to mediate with explanatory

theories between the opposing conceptions of Heraclitus and Par-

menides.

Out of these needs arose, about the middle of the fifth century,

side by side, and with many reciprocal relations, positive and polem-

ical, the theories of Empedodes, Anaxagoras, and Leucippus, founder

of the Atomistic School of Abdera. The number of these theories

and their well-known dependence upon one another prove that in

spite of the distance by which individual men and schools found

themselves separated, there was already a great vigour in exchange

of thought and in literary activity. The picture of this life takes

on a much fuller form as we reflect that tradition, in sifting its

material, has obviously preserved only the memory of what was

most important, and that each of the names remaining known to

us indicates, in truth, an entire circle of scientific activity.

The Pythagoreans, during this same period, occupied a peculiar

position at one side. They also took up the metaphysical problem

given by the opposition between Heraclitus and the Eleatics, but

hoped to find its solution by the aid of mathematics, and, by their

theory of numbers, as whose first literary representative Philolaus is

known, added a number of most important factors to the further

movement of thought. The original purpose or tendency of their

league made itself felt in their doctrines, in that, in fixing these,

they conceded a considerable influence to considerations of (ethical

or aesthetic) worth. They indeed attempted a scientific treatment

of ethical questions as little as did the entire philosophy of this

period, but the cosmology which they based upon their astronomical

ideas, already widely developed with the help of mathematics, is

yet at the same time permeated by aesthetic and ethical motives.

Of the Milesian School only three names— Thales, Anaximander, and An-aximenes— have been handed down to us. From this it appears that the school

flourished in what was then the Ionic capital during the entire sixth century,

and perished with the city itself, which was laid waste by the Persians in 494,

after the battle of Lade.Thales, sprung from an old merchant family, is said to have predicted the

solar eclipse in 585, and survived the invasion of the Persians in the middle of

the sixth century. He had perhaps seen Egypt, and was not deficient in mathe-matical and physical knowledge. So early an author as Aristotle did not knowwritings from him.Anaximander seems to have been little younger. Of his treatise irepl <pv<reus

a curious fragment only is preserved. Of. Neuhäuser (Bonn, 1883). — Biisgen,

Ueber das diretpov des A. (Wiesbaden, 1807).It is difficult to determine the period of Anaximenes. It falls probably about

5G0-500. Almost nothing of his work irepl <puaeus remains.Aside from that given by Aristotle (in the beginning of the Metaphysics) we

owe our meagre information concerning the theories of the Milesians chiefly to

the Commentary of Simplicius. Of. H. Kitter, Geschichte der jonischen Philos-

ophie (Berlin, 1821) ; R. Seydel, Der Fortschritt der Metaphysik unter den ältes-

ten jonischen Philosophen (Leips. 1861).

Page 50: Windelband, History of Philosophy, Vol. 1

30 The Grreeks: Cosmological Period. [Part I.

At the head of the Eleatic School, Xenophanes, who at ail events wasconcerned in its establishment, is generally placed. Born about 570 in Colophon,he fled in 546, in consequence of the Persian conquest of Ionia, and gained aliving as wandering poet. At last, in Elea, founded by the Ionians who fled into

Magna Grsecia, he found a permanent dwelling. He died after 480. The frag-

ments of his partly gnomic, partly philosophical, sayings have been collected byKarsten (Amsterdam, 1835). Concerning him see Fr. Kern (Naumburg, 1864,

Oldenburg, 1867, Danzig, 1871, Stettin, 1874 and 1877) and J. Freudenthal (Bres-

lau, 1886).

Farmenides, an Eleatic of renowned family, who was not a stranger to thePythagorean society, wrote about 470. The fragments of his didactic poemhave been collected by Peyron (Leips. 1810) and H. Stein (Leips. 1864). [Met.tr. in Jour. Spec. Phil., IV.] The lost treatise of Zeno (about 490-430) wasprobably the first which was separated into chapters and arranged dialectically.

He, too, came from Elea.

Meliasos, on the contrary, was the Samian general who conquered the Athe-nians in 442. Concerning his personal connection with the Eleatic school nothingis Known. A. Fabst, De M. Fragmentis ^Bonn, lbb9).

The unimportant fragments of the Eleatics are in a measure supplemented bythe accounts of Aristotle, Simplicius, and others. The pseudoAristotelian work,De Xenephone, Zenone, Gorgia (Arist., Berl. ed., 974 ff.), which must be usedwith great discretion, gives an account in the first chapter probably of Melissos •

in the second, from confusedly intermingling sources, of Zeno ; in the third, of

Gorgias.

Heraclitus of Ephesus ("the Obscure"), about 536^470, disgusted with theever-growing power of the democracy, gave up the high position which was his

by birth, and in the moody leisure of the last decade of his life, wrote a treatise

which was pronounced difficult of comprehension even by the ancients, whilethe fragments of it which we possess are often very ambiguous. Collected andedited by P. Schuster (Leips. 1873) and J. Bywater (Oxford, 1877). Cf. Fr.Schleiermacher {Ges. W-, III. Abth., Bd. 2, pp. 1-146); J. Bernays (Ges. Abhand-lungen, Bd. I., 1885); F. Lasalle (2 Bde., Berlin, 1858); E. Pfleiderer (Berlin,

1886). [G. T. W. Patrick, Heraclitus in Am. Jour. Psy., L, 1888, contains trans,

of the Fr.]The first Dorian in the history of philosophy is Empedocles of Agrigentum,

about 490-430, a priestly and prophetic personality, much regarded in his char-

acter as statesman, physician, and worker of miracles. He had, too, relations

with the Sicilian school of orators, of which the names of Korax and Tisias arefamiliar ; and besides his Ka9a.pfj.ol (Songs of Purification) has left a didactic

poem, the fragments of which have been published by Sturz (Leips. 1805),Karsten (Amsterdam, 1838), and Stein (Bonn, 1852).Anaxagoras of Klazomene (500 till after 430) settled, toward the middle

of the fifth century, in Athens, where he made friends with Pericles. In 434he was accused of impiety and obliged to leave the city, and founded a schoolin Lampsacus. Schaubach (Leips. 1827) and Schorn (Bonn, 1829) have col-

lected the fragments of his treatise, wepl <pvo-ews. Cf. Breier (Berlin, 1840),Zevort (Paris, 1843).

So little is known of the personality of Leucippu3, that even in ancienttimes his very existence was doubted. The great development of the atomistictheory by Democritus (see ch. 3) had completely overshadowed its founder.But traces of Atomism are to be recognised with certainty in the entire structureof thought after Parmenides. Leucippus, if not born in Abdera, yet activethere as head of the school out of which Protagoras and Democritus went later,

must have been contemporary with Empedocles and Anaxagoras, even thoughsomewhat older. Whether he wrote anything is uncertain. Cf. Diels, Verh.der Stett. Philol. Vers. (1886). —A Brieger, Die Urbewegung der Atome (Halle,

1884); H. Liepmann, Die Mechanik der leucipp-demokritischen Atome (Leips.

1885).The Pythagorean Society first appeared in the cities of Magna Graecia as

a religious-political association toward the end of the sixth century. Its founderwas Pythagoras, of Samos, who, born about 580, after long journeys, whichprobably led him toward Egypt also, made the aristocratic city of Crotona the

starting-point of a reform movement which had for its aim a moral and religious

Page 51: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 4.] Conceptions of Being. 31

purification. We are first apprised of the internal relations of the society

through subsequent narratives (Jamblichus, De Vita Pythagorica, and l'orphyrius,

De Vita Pythagorce published by Kiesling (Leips. 1815-1(5), whose trustworthinessis doubtful. It seems, however, to be certain that already the old society imposeddefinite duties upon its members, even for private life, and introduced the prac-

tice of working in common at intellectual pursuits, especially at music andmathematics. In consequence of its political position (in regard to whichB. Krische, Göttingen, 1830) the external conditions of the society assumed at

first a very favourable form, inasmuch as, after the plunder of the democraticSybaris, 509, Crotona won a kind of hegemonic influence in Magna Grtecia.

In time, however, the Pythagoreans became the losers in the bitter partystruggles of the cities, and often suffered bitter persecution, by which thesociety was finally destroyed in the fourth century.

To Pythagoras himself, who died about 500, we can trace back no philosoph-

ical writings, although the subsequent myth-making process sought so strenu-

ously to make him the idol of all Hellenic wisdom. (E. Zeller in Vortr. u.

Abhandl., L, Leips. 1865.) Plato and Aristotle knew only of a philosophy of

the Puthagoreans. Philolaus, who seems to have been somewhat younger thanEmpedocles and Anaxagoras, appears as the most prominent representative of

this philosophy. Almost nothing is known of the circumstances of his life, andthe fragments of his treatise (ed. by Boeckh, Berlin, 1819 ; cf. C. Schaar-schmidt, Bonn, 1864) lie under considerable suspicion.

Of the remaining adherents of the society, only the names are known. Thelatest representatives came into so close relations with the Platonic Academythat, as regards their philosophy, they may almost be said to have belonged to

it. Among them Archytas of Tarentum, the wsll-known savant and statesman,

should be mentioned. Concerning the very doubtful fragments attributed to

him, cf. G. Hartenstein (Leips. 1833), Fr. Petersen (Zeitschr. f. Alterthumsk

;

1836), O. Gruppe (Berlin, 1840), Fr. Beckman (Berlin, 1844).

The reports concerning the teaching of the Pythagoreans, especially in the later

accounts, are clouded by so many additions from foreign sources, that perhapsat no point in ancient philosophy is it so difficult to determine the actual facts

in the case as here, even if we sift out the most trustworthy, namely Aristotle

and his best taught commentators, notably Simplicius, many dark points andcontradictory statements remain, particularly in details. The reason for this

lies probably in the fact that in the school, which for a time was widely extended,

various trends of thought ran side by side, and that among these the general fun-

damental thought first brought forward perhaps by Philolaus, was worked out

in different ways. It would be of great service to attempt such a separation.

H. Ritter, Geschichte der pythagoreischen Philosophie (Hamburg, 1826) ;

Rothenbücher, Das System der Pythagoreer nach Aristoteles (Berlin, 1867) ;

E. Chaignet, Pythagore et la Philosophie pythagoricienne (2 vols., Paris,

1873).

§ 4. The Conceptions of Being.

The fact that things of experience change into one another was

the stimulus to the first philosophical reflections, and wonder 1 at

this must indeed have arisen early among a people so mobile and

with so varied an experience of Nature as the Ionians. To this

fact, which furnished the fundamental motive of its reflection, the

Ionic philosophy gave liveliest expression in Heraclitus, who seems

to have been unwearied 2 in seeking the most pointed formulations

for this universal mutability of all things, and especially for the

sudden changes of opposites into each other. But while myth gave

x Cf. upon the philosophical value of the eavfxdiCLv, Arist. Met. I. 2, 982 b 12.

2 Fragm. (Schust.) 41-44, 60, 63, 67.

Page 52: Windelband, History of Philosophy, Vol. 1

32 The Greeks : Cosmological Period. [Part I.

to this view the garb of a fabled, account of the formation of the

world, science asked for the abiding ground of all these changes,

and fixed this question in the conception of the cosmic matter, or

" ivorld-stuff " (Weltstoff), which experiences all these transforma-

tions, from which all individual things arise, and into which they

become again transformed (dpx7?)- ^n this conception x was tacitly-

contained the presupposition of the unity of the world; whether the

Milesians 2 already sought to justify this we do not know. It was a

later eclectic straggler 3 who first attempted to justify this Monism

by the transformation of all things into one another, and by the

inter-connection of all things without exception.

1. That, however, a single cosmic matter, or world-stuff, lies at

the basis of the entire process of nature, appears in ancient tradi-

tion as a self-evident presupposition of the Ionic School. The only

question was to determine what this elementary matter was. The

nearest course was then to seek for it in what was given in experi-

ence, and so Thales declared it to be ivater; Anaximenes, air. To

this choice they were probably determined only by the mobility,

changeability, and apparent inner vitality 4 of water and air. It is

evident, too, that the Milesians thought little in this connection of

the chemical peculiarities of water and air, but only of the states

of aggregation 5 concerned. While the solid appears in itself dead,

moved only from without, the liquid and volatile make the impres-

sion of independent mobility and vitality ; and the monistic prepos-

session of this first philosophising was so great that the Milesians

never once thought of asking for a reason or ground of this cease-

less change of the cosmic matter, but instead assumed this as a self-

intelligible fact— a matter of course— as they did all change or

occurrence ; at most they described its individual forms. The cos-

mic matter passed with them for something in itself living : they

thought of it as animated, just as are particular organisms,6 and for

this reason their doctrine is usually characterised from the stand-

point of the later separation in conceptions as Hylozoism.

1 Which Aristotle in the Met. I. 3, 983 b 8, has denned, not without theadmixture of his own categories.

2 The expression äpxVi which, moreover, bears in itself the memory of the

chronological fancies of the Cosmologists, is said by Simplicius to have beenused first by Anaximander.

3 Diogenes of Apollonia. Cf. Simpl. Phys. (D.) 32 r 151, 30, and Arist. Gen. et

Corr. I. 6, 322 b 13.4 Schol. in Arist. 514 a 33.5 For vSup, vypöv is frequently substituted. With regard to the drip of Anaxi-

menes the accounts are such that the attempt has been made to distinguish his

metaphysical "air" from the empirical: Ritter, I. 217; Brandis, I. 144.6 Plut. Plac. I. 3 (Doxogr. D. 278). Perhaps this is intended in the conjec-

ture of Aristotle, Met. I. 3, 983 b 22.

Page 53: Windelband, History of Philosophy, Vol. 1

Jhap. 1, § 4.] Conceptions of Being : The Milesians. 38

2. If we ask, however, why Anaximenes, whose doctrine, like

that of Thales, seems to have kept within the bounds of experience,

substituted air for water, we learn 1 that he believed air to have a

characteristic which water lacked, — a characteristic, too, which his

predecessor Anaximander had postulated as indispensable for the

conception of primitive matter, viz. that of infinity. As motive for

this postulate of Anaximander there is related the argument that a

finite cosmic matter would exhaust itself in the ceaseless succession

of productions. 2 But Anaximander had also seen that this demandmade by the conception of the apxn could not be satisfied by any

matter or substance which we can perceive, and had on this account

transferred the cosmic matter beyond experience. He maintained

boldly the reality of an original ground of things, possessing all the

properties that are necessary, if we are to derive the changes in the

world of experience from something itself abiding and raised above

change, — even though such a ground might not be found in experi-

ence. He drew from the conception of the apxv the consequence,

that though no object of experience corresponds to this conception,

we must yet, to explain experience, assume such a conception behind

it as real and conditioning it. He therefore called the cosmic mat-

ter "the Infinite'''' (to awapov), and ascribed to it all the qualities

postulated in the conception of the apxq : that is, that it had never

begun to be, and was imperishable, inexhaustible, and indestructible.

The conception of matter, thus constructed by Anaximander is,

nevertheless, clear only in the respect that it is to unite within it

spatial infinity and the quality of being without beginning or end

in time, and thus the mark of the all-embracing and all-determim

ing; 3 on the other hand, with reference to its qualitative deter-

mination, it cannot be made clear what the philosopher intended.

Later accounts give us to understand that he expressly maintained

that the original matter was qualitatively undetermined or indefinite

(dopio-Tos),4 while the statements of Aristotle 5 speak more for the

assumption of a mixture of all kinds of matter known in experience,

— a mixture completely adjusted or equalised, and therefore as a

whole indifferent or neutral. The most probable view here is, that

Anaximander reproduced in the form of an abstract conception the

1 Simpl. Phys. (D.) 6' 24, 26.2 Plut. Plac. I. 3 (Doxogr. D. 277) ; Arist. Phys. III. 8, 208 a 8.

8 Arist. Phys. III. 4, 203 b 7.

* Schol. in Arist. 614 a 33 ; Herbart, Einleitung in die Philosophie (Ges.

W., I. 196).6 Met. XII. 2, 1069 b 18, and especially Phys. I. 4, 187 a 20. Cf. also Simpl.

Phys. (D.) 33 r 154, 14 (according to Theophrastus). This much-treated contro-

versy will be spoken of more in detail below (§ 6).

Page 54: Windelband, History of Philosophy, Vol. 1

34 The Greeks : Gosmological Period. [Part. L

unclear idea of the mythical chaos which was "one" and yet also

"all." This he did by assuming as the cosmic matter an infinite,

corporeal mass, in which the various empirical substances were so

mixed that no definite quality could be ascribed to it as a whole.

For this reason, however, the separation of the individual qualities

out of this self-moved matter could no longer be regarded as properly

a qualitative change in it. With this view the conception of the

unity of the world as regards quality would be given up, to be sure,

and an essential preparation made for the later development.

3. Still another predicate was given by Anaximander to the In-

finite,— to Odov, the divine. As a last remembrance of the religious

home in which scientific reflection arose, it shows for the first time

the inclination of philosophers, constantly recurring in history, to

view as " Deity " the highest conception which theory has led them

to use for explaining the world, and so to give it at the same time

a sanction for the religious consciousness. Anaximander's matter is

the first philosophic conception of God, the first attempt, and one

which remains still entirely within the physical, to strip the idea

of God of all mythical form.

But while the religious need thus maintained itself in the deter-

mination of metaphysical conception, the possibility of an influence

of the results of science upon the religious life was brought nearer, the

more these results met and responded to an impulse which hitherto

had been dominant only in an obscure and uncertain manner within

that life. The transformation which the Greek myths had undergone,

as well in the import given them in cosmogonic fancy as in that given

to their ethical interpretation, tended everywhere toward a mono-

theistic culmination (Pherecydes, Solon) ; and to this movement

its final result, a clearly outspoken monism, was now proffered by

science.

This relation was brought to expression by Xenophanes, not a

thinker and investigator, but an imaginative disciple of science,

strong in his convictions, who brought the new teaching from East

to West and gave it a thoroughly religious colouring. His mainte-

nance of monotheism, which he expressed as enthusiastic intuition in

the saying, 1 that whithersoever he looked all was constantly flowing

together for him into one Nature (/u,iav eis </>vcriv), took on at once,

however, that sharp polemic turn against the popular faith, by which

he is principally characterised in literature. The scorn, which he

poured out with abundant wit over the anthropomorphism of myth-

ology, 2 the anger with which he pursued the poets as the portrayers

i Timon in Sext. Emp. Pyrrh. Hyp. I. 224. 2 Clem. Alex. Strom. V. 601.

Page 55: Windelband, History of Philosophy, Vol. 1

Chai\ 1. § 4.] Conceptions of Being : Xenophanes. 35

of these divine figures provided with all the weaknesses and vices of

human nature, 1— these rest upon an ideal of God which will have

the Supreme Being regarded as incomparable with man in both

bodily and mental characteristics. When he passes to positive at-

tributes, Xenophanes becomes more obscure. On the one hand, the

deity as ev koX ttov is identified with the universe, and to this " World-

God " are then ascribed all the predicates of the Milesian dpxv

(eternity, existence that has not become what it is, imperishability);

on the other hand, qualities are ascribed to the deity, some of which

are spatial, as the spherical form, while others are psychical func-

tions. Among these latter the omnipresence of the knowing activity

and of the rational guidance of things is expressly mentioned. In

this respect the World-God of Xenophanes appears only as the

highest among the rest of " gods and men."

While here a predominantly theological turn of philosophy is

already manifested, the exchange of the point of view of metaphysics

and natural science taken by Anaximander, for the religious point

of view of Xenophanes shows itself in two essential deviations.

The conception of the World-God is for the latter an object of

religious reverence, and scarcely a means for understanding Nature.

The Colophonian's sense for knowledge of Nature is slight, his ideas

are in part very childlike, and, as compared with those of the Mile-

sians, undeveloped. And so for his views, the characteristic of

infinity, which Milesian science regarded as necessary in the cosmic

matter, could be dispensed with ; on the contrary, it seemed to him

more in accordance with the dignity of the divine Nature, 2 to think

of this as limited within itself, as entirely shut up or complete, con-

sequently as regards its spatial aspect, spherical. And while the

Milesians thought of the original ground of things as ever in motion

spontaneously, and as characterised by living variety in its inter-

nal structure, Xenophanes struck out this postulate hitherto in use

for the explanation of Nature, and declared the World-God to be

immovable and perfectly homogeneous in all its parts. How, indeed,

he thought that the variety of individual things whose reality he

did not doubt, could be reconciled with this view, must remain

uncertain.

4. As was required by the conception of change, the Milesian

conception of the World-substance had united without clear discrim-

ination two essential elements : the one that of a substance re-

maining like itself, the other that of independent or self-subsistent

1 Sext. Emp. Adv. Math. IX. 19.3.2 Hippol. Ref. I. 14 (Doxogr. D. 505). In other passages, again, it is said

Shat he would have the deity thought neither limited nor unlimited (?>.

Page 56: Windelband, History of Philosophy, Vol. 1

36 The Greeks : Cosmological Period. [Pakt I.

cnangeability. In the thought of Xenophanes the first element wasisolated ; the same process took place for the second through Hera-

clitus. His doctrine presupposes the work of the Milesians, from

the conclusion of which it is separated by a generation, in this way

:

their effort to determine or define in conceptions an abiding world-

ground has been recognised as hopeless. There is nothing abiding,

either in the world or in its constitution taken as a whole. Notonly individual things, but also the universe as a whole, are involved

in perpetual, ceaseless revolution: all flows, and nothing abides. Wecannot say of things that they are ; they become only, and pass away

in the ever-changing play of the movement of the universe. That,

then, which abides and deserves the name of deity, is not a thing,

and not substance or matter, but motion, the cosmic process, Becom-

ing itself.

To meet a strong demand that seems made by this turn to abstrac-

tion, Heraclitus found help in the sensuous perception in which this

motion presented itself to him : that of fire. The co-operation of

this in the conversion of things of Nature into each other had been

already noticed by the Milesians ; to this may have been added

ancient Oriental mystical ideas, which contact with the Persians

made especially accessible to the Ionians of that day. But whenHeraclitus declared the world to be an ever-living fire, and Fire,

therefore, to be the essence of all things, he understood by this apxq

not a material or substance which survived all its transformations,

but just the transforming process itself in its ever-darting, vibrating

activity {züngelnde), the soaring up and vanishing which corre-

spond to the Becoming and passing away. 1

At the same time, however, this idea takes on a still firmer form,

in that Heraclitus emphasised much more strongly than the Mile-

sians the fact that this change is accomplished in accordance with

definite relations, and in a succession that remains always the same. 2

This rhythm of events (which later times have called the uniformity

of Nature under law) is therefore the only permanent ; it is termed

by Heraclitus the destiny {el/xapfxevr)), the order (6Ykt/), the reason

(Aoyos) of the world. These predicates, in which physical, ethical,

1 The difficulty of ascribing to such a motion without any substrate, to a mereBecoming, the highest reality and the capacity to produce things, was evidentlyvery much less for undeveloped thought not yet conscious of its categories thanfor later apprehension. The conception of Becoming as fire, hovering betweenthe symbolic and the real meaning of the term, was supported by the use of

language which treats of functions and relations as also substantives. ButHeraclitus does not disdain to let the dim idea of a World-substance stand in thebackground in his metaphors (of the clay kneaded ever anew, of the drinkcontinually stirred).

2 Further in detail on this point in the following section.

Page 57: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 4.] Conceptions of Being : Heraclitus, Parmenides. 37

and logical order in the world appear as still identified, prove only

the undeveloped state of thought which does not yet know how to

separate the different motives. The conception, however, which

Heraclitus has grasped with complete clearness, and carried though

with all the strength of his austere personality, is that of order, a

conception, nevertheless, whose validity was for him as much a

matter of conviction as of knowledge.

5. In evident opposition to this theory of the Ephesian, the con-

ception of Being was worked out by Parmenides, the head of the

Eleatic School, and the most important thinker of this period. Yet

it is not easy to reconstruct his formulation of this conception from

the few fragments of his didactic poem, the quite unique character

of which consists in the union of dryest abstraction with grand and

rich imagery. That there is a Being («fo-n yap eirai), is for the Ele-

atic a postulate of such cogent evidence that he only states this

position without proving it, and that he explains it only by a nega-

tive turn of thought which first discloses to us completely the sense

in which we are to understand his main thought. " Non-being "

(firj elvai), he adds, or that which "is" not (to /a?) iov), cannot be

and cannot be thought. For all thought is in relation to a some-

thing that is, which forms its content. 1 This view of the correla-

tive nature of Being and consciousness leads so far with Parmenides

that the two, thought and Being, are declared to be fully identical.

No thought to whose content Being does not belong,— no Being

that is not thought : thought and Being are the same.

These propositions, which look so abstractly ontological if we con-

sider only the words, take on quite another meaning when we con-

sider that the fragments of the great Elean leave no doubt as to

what he desired to have regarded as " Being" or that which "is."

This was corporeality, materiality (rb ir\£ov). For him, "being" and" filling space " are the same. This " Being," this function of filling

space, is precisely the same in the case of all that "is"; there is,

therefore, only the one, single Being which has no internal distinc-

tions. " Non-being," or what is not [has not the attribute of Being],

means, accordingly, incorporeality, empty space (to kcvov). This

double meaning of the elvai (Being) employed by Parmenides, ac-

cording to which the word means at one time " the full " and at an-

other time " Keality," leads then to the proposition that emjyty space

cannot be.

Now for the na'ive, sensuous way of looking at tilings which

lurks even in these principles of Parmenides, the separateness of

1 Fr., ed. Karsten, vv. 9± ff.

Page 58: Windelband, History of Philosophy, Vol. 1

38 The Crreeks : Cosmologieal Period. [Part I

things, by virtue of which they present themselves in their plurality

and multiplicity, consists in their separation by empty space ; and,

on the other hand, all that takes place in the corporeal world, i.e.

all motion, consists in the change of place which the " full " experi-

ences in the "empty" (or the "Void"). If, therefore, the Void is

not real or actual, then the plurality and motion of individual things

cannot be real.

The number and variety of things presented in co-existence and

succession by experience had given the Milesians occasion to ask

for the common abiding ground of which all these things were

metamorphoses. When, however, the conception of cosmic sub-

stance or world-stuff has culminated with Parmenides in the con-

ception of Being, there seems so little possibility of uniting these

individual things with it, that reality is denied them, and the one

unitary Being remains also the only being. 1 The conception formed

for the purpose of explanation has so developed internally that to

maintain it involves the denial of that which was to be explained

by it. In this sense the Eleatic doctrine is acosmism : the mani-

foldness of things has sunk in the All-one : the latter alone " is,"

the former are deception and seeming.

According to Parmenides, however, we are to predicate of the

One that it is eternal, has never come into being, is imperishable,

and especially (as Xenophanes had maintained) that it is through

and through one in kind, one with itself, without any distinctions

or differences, i.e. completely homogeneous and absolutely unchange-

able. He follows Xenophanes also in regarding the One as limited,

complete, and definitive. Being is then a well-rounded sphere, per-

fectly homogeneous within itself, and this only and unitary world-

body is at the same time the tvorld-thought, 2 simple, excluding all

particulars from itself : to yap irXeov ecrTi vorjfia.

6. All these attempts, in part fantastic, in part regardlessly

abstract, were needed in order to gain the presuppositions for the

development of the first usable conceptions for apprehending Nature.

For important as were the motives of thought that had come to

recognition therein, neither the world-stuff or cosmic matter of the

Milesians, nor the "Fire-Becoming" of Heraclitus, nor the Being of

Parmenides were available for explaining Nature. Now the imper-

fection of the first had become clear through the contrast which

1 A great role in these considerations of the Eleatics is obviously played bythe ambiguities in language, by which, on the one hand, the ev means bothnumerical unity and also qualitative unity or simplicity, while the verb elvai hasnot only the function of the copula, but also the meaning of " Reality."

2 Hence, terms like " materialism " and " idealism " do not apply to this naive'•lentification of consciousness and its object, the corporeal world.

Page 59: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 4.] Conceptions of Being : Empedocles. S9

separated the two latter as by a gulf, and with the recognition of

this, occasion was given for the more independent investigators of

the next period to separate in their conceptions the two motifs

(being and becoming), and by setting them over against one another

to think out new forms of relation, out of which permanently valua-

ble categories for the knowledge of Nature resulted.

These mediating attempts have in common, on the one hand, the

recognition of the Eleatic postulate that that which " is " must be

thought throughout not only as eternal, without a beginning and

imperishable, but also as homogeneous, and as regards its qualities

unchangeable ; on the other hand, however, they assent also to the

thought of Heraclitus that an undeniable reality belongs to Becom-

ing and change {Geschehen), and so to the manifoldness of things.

Common to them, also, in their adjustment of these two needs of

thought is the attempt to assume a plurality of beings, each of which

should satisfy for itself the postulate of Parmenides ; while, on

the other hand, by changing their spatial relations, they were to

bring about the changeful variety of individual things which expe-

rience shows. If the Milesians had spoken of qualitative changes

of the cosmic substance or matter, the Eleatic principle had ex-

cluded the possibility of it ; if, nevertheless, change ought to receive

recognition, as with Heraclitus, and be attributed to Being itself,

it must be reduced to a kind of change which leaves untouched

the qualities of the existent. Such a change, however, was think-

able only as a change of place, i.e. as motion. The investigators of

Nature in the fifth century maintained, therefore, with the Eleatics,

the (qualitative) unchangeableness of the existent, but against the

Eleatics, its plurality and motion ;* with Heraclitus, they insisted

upon the reality of occurrence and change, and against Heraclitus,

upon the Being of permanent and unchangeable substances as under-

lying and producing the same. Their common view is this : there

is a plurality of existing beings which, unchangeable in them-

selves, make the change and variety of individual things compre-

hensible.

7. This principle seems to have been asserted first and in its

most imperfect form by Empedocles, — in a form, however, that was

widely influential historically. He put forward as " elements" 2 the

four which are still current in the popular modes of thought,— earth,

1 Later (Plato, Theae.t. 181 D ; Arist. var. loc), dXXoWts (qualitative change)and irepupopd (change of place) are contrasted as species of Kivrjais or neTaßoXJ/.

In reality this is done here, though the terms are as yet lacking.2 Instead of the later expression o-toix««*, we find in Empedocles the more

poetic term " roots of all things," pi^nara.

Page 60: Windelband, History of Philosophy, Vol. 1

40 The G reeks : Cosmological Period. [Part L

water, air, and fire.1 Each of these is according to this system,

without beginning and imperishable, homogeneous and unchange-

able, but at the same time divisible into parts, and in these parts

capable of change of place. Out of the mixture of the elements

arise individual things, which in turn cease to exist when the mix-

ture is separated into the elements ; to the kind of mixture made

are due the various qualities of individual things, which are often

different from the properties of the elements themselves.

At the same time the note of unchangeableness and a deviation

from the Milesian Hylozoism assert themselves in the system of

Empedocles to the extent that he could not assign independent ca-

pacity of motion to these material elements which experience only

changing states of motion and mechanical mixings. On this account

he was obliged to seek a catcse of motion independent of the four

elements. As such a cause he designated love and hate. The out-

come, however, of this first attempt to set over against a dead matter,

deprived by abstraction of all motion of its own, the force which

moves it, as a metaphysically independent something, was very

obscure. Love and hate are, with Empedocles, not mere properties,

functions, or relations of the elements, but rather independent

powers set over against them ; but how we are to think the reality

of these moving forces is not disclosed in any satisfactory way in the

fragments. 2 Only this seems certain, that in fixing the dual nature

of the principle of motion the thought was also operative that two

distinct causes, love and hate, were requisite to account for the

good and the evil in the change of things of our experience,3— a first

indication that determinations of " worth " or value are beginning

to be introduced into the theory of Nature.

8. Empedocles thought it possible to derive the special qualities

of individual things from the proper mixture of the four elements :

whether he attempted so to derive them, and if so, how, we do not

indeed know. This difficulty was avoided by Anaxagoras, who,

from the Eleatic principle that nothing that is can arise or pass

away, drew the conclusion that as many elements must be assumed

1 Aside from dependence upon his predecessors, his selection was evidentlydue to the inclination to regard the different states of aggregation as the origina:

essence of things. No importance seems to have attached to the number four,

in this. The dialectical construction which Plato and Aristotle gave for this is

quite remote from the thought of the Agrigentine.2 If <pt\ia and veiKos are occasionally counted by the later recorders as fifth

and sixth äpxv of Empedocles, we must not infer from this that he regardedthem as substances. His obscure and almost mythical terminology rests, for

the most part, upon the fact that conceptions standing for functions are substan-tives in language. 3 Arist. Met. I. 4, 984 b 32.

4 He called them criripfiaTa (seeds of things), or also simply xP^ilJ-aTa (sub-

stances).

Page 61: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 4.] Conceptions of Being : Anaxagoras. 41

as there are simple substances in the things of experience, meaning

by simple substances those which on repeated division always sep-

arate into parts qualitatively the same with their wholes. Such

elementary substances were later, in accordance with his definition,

called homoiomeriai. At that time, however, when only mechanical

division or change of temperature were known as means of investi-

gation, this conception of element (in principle entirely correspond-

ing to the conceptions of the chemistry of to-day) applied to the

greater part of the substances given in experience, 1 and on that ac-

count Anaxagoras maintained that there were countless elements dif-

fering in form, colour, and taste. He held that they were present

throughout the entire universe in a very finely divided state. Their

coming together or compounding (o-vy/coto-is) constitutes the arising,

their separation (SiaKpicris) the passing away, of individual things.

There is, accordingly, something of every substance present in every-

thing: it is only for our sensuous apprehension that the individual

thing takes on the properties of that substance or of those sub-

stances which may be present in a preponderating degree.

The elements, as the true being, are regarded now by Anaxagoras

also as eternal, without beginning or end, unchangeable, and though

movable in space, yet not in motion of themselves. Here, too, then,

we must ask for a force which is the cause of motion. Since, how-

3ver, this force must be regarded as existent, a something that is,

A-naxagoras hit upon the expedient of assigning it to a special,

«single sort of matter or elementary substance. This force-element

or motive-matter {Bewegungsstoff) is conceived to be the lightest and

most mobile of all elements. In distinction from all the others it is

that one of the homoiomeriai which alone is in motion of itself, and

communicates this its own motion to the rest ; it moves itself and

the rest. To determine the inner nature of this " force-substance,"

however, two lines of thought unite : the property of originating mo-

tion is, for the naive mode of looking at things, the surest sign of the

animate; this exceptional kind of matter, then, which is self-moved,

must be animate matter or "soul-stuff" (Seelenstoff), its quality

must be animate or psychical. 2 And, secondly, a power is knownthrough its effect : if, now, this motive-matter is the cause of the

formation of the world, to bring about which it has separated out

the remaining idle elements, then we must be able to know its

nature from this which it has accomplished. But the universe, in

particular the regular revolution of the stars, makes the impression

1 According to the fragments of Anaxagoras, bones, flesh, and marrow also;

on the other hand, the metals.2 [The Greek ^vxv and German Seele include both these meanings.]

Page 62: Windelband, History of Philosophy, Vol. 1

42 The Greeks : Cosmoiogical Period. [Part L

of beautiful and purposive order (xooytos). Such a mastering of

gigantic masses in a harmonious system,— this undisturbed circling

of countless worlds, on which Anaxagoras turned his wondering

contemplation, it seemed to him could be the result only of a mind

arranging the movements according to ends, and ruling them. For

this reason he characterised the force-substance as Reason (vovs) or

as " Thought-stuff."

The voGs of Anaxagoras is then a stuff or substance, a corporeal

element, homogeneous, unproduced, aud imperishable, diffused in a

finely divided state throughout the universe ; different from the

other substances, however, not only in degree, as being the finest,

lightest, and most mobile, but also in essence, since it alone is self-

moved, and by virtue of its own motion moves the other elements in

the purposive way which we recognise in the order of the world.

This emphasising of the order in the universe is a Heraclitic element

in the teaching of Anaxagoras, and the conclusion drawn from the

ordered movements to a rational cause of them, acting according to

ends, is the first instance of the teleological explanation of nature. 1

With this procedure a conception of worth ( Werthbegriff)— namely,

oeauty and perfection— is made a principle of explanation in the

theoretical field also.

9. The Atomism of Leucippus developed from the Eleatic concep-

tion of Being in a direction opposite to that just traced. While

Empedocles maintained that some, and Anaxagoras that all, qualities

were metaphysically primitive, the founder of the school of Abdera

remained in accord with the position of Parmenides, that no "Being"

belongs to any of all the various qualitative determinations exhibited

by experience, and that the sole property of Being is the property of

filling space, corporeality, to 7rAe'ov. If now, however, the plurality of

things, and the mutations taking place among them as they come

and go, were to be made intelligible, then instead of the single world-

body, with no internal distinctions which Parmenides had taught, a

plurality of such must be assumed, separated from one another, not

by other Being, but by that which is not Being, Non-being : i.e. by the

incorporeal, by empty space. This entity, then, which is Non-being [i.e.

not Being in the true sense], must have in its turn a kind of Being,

or of metaphysical reality ascribed to it,2 and Leucippus regarded it

1 As such he was praised by Plato (Phced. 97 B), and overestimated byAristotle (Met. I. 3, 984 b). Cf., however, § 5. The moderns (Hegel) haveadded the further over-estimate of seeking to interpret the vovs as an immate-rial principle. But the fragments (Simpl. Phys. (D.) 33v 156, 13) leave nodoubt that this lightest, purest element, which does not mingle with the rest,

but only plays about them and moves them as living force, was also a space-

filling matter or stuff. 2 Plut. Adv. Col. 4, 2, 1109.

Page 63: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 4.] Conceptions of Being : Leucippus, Zeno. 43

as the unlimited, the a-rrupov, in contrast with the limitation which

Being proper possesses, according to Parmenides. Leucippus, there-

fore, shatters in pieces the world-body of Parmenides, and scatters

its parts through infinite space. Each of these parts, however, is,

like the absolute Being of Parmenides, eternal and unchangeable,

without beginning, indestructible, homogeneous, limited, and indi-

visible. Hence these portions of Being are called atoms, ärofxoi;

and for the reasons which had led Anaximander to his concept

of the airupov Leucippus maintained that there were countless

numbers of such atoms, infinitely varied in form. Their size must

be taken as imperceptibly small, since all things in our experience

are divisible. Since, however, they all possess only the one like

quality of filling space, differences between them can be only quan-

titative ; differences in size, form, and situation.

Out of such metaphysical considerations grew the concept of the

atom, which has proved so fruitful for the theoretical science of

Nature just because, as was evident already in the system of Leu-

cippus, it contains the postulate that all qualitative differences

exhibited by Nature are to be reduced to quantitative. The things

which we perceive, Leucippus taught, are combinations of atoms

;

they arise when atoms unite, and pass away when they part. The

properties which we perceive in these complexes are only seeming

or appearance ; there exist in truth only the determinations of size,

form, arrangement, and situation of the individual atoms which

constitute Being.

Empty space is, accordingly, the presupposition as well for the

uniting and separating of atoms as for their separateness and shape.

All " becoming," or change, is in its essence motion of atoms in space.

If we ask for the ground of this motion of the atoms, 1 since space

as properly not a true Being cannot be allowed as cause, and

Atomism recognises nothing as actual except space and the atoms,

this ground can be sought only in the atoms themselves; i.e. the

atoms are of themselves in motion, and this, their independent mo-

tion, is as truly without beginning and end as is their being. And as

the atonu are indefinitely varied in size and form, and completely

independent of one another, so their original motions are infinite in

variety. They fly confusedly about in infinite space, which knows

no above and below, no within and without, each for itself, until

their accidental meeting leads to the formation of things and worlds.

The separation between the conceptions of matter and moving force

1 Arist. Phys. VIII. 1, 252 a 32, says of the Atomists that they did not ask as

to the origin of motion— as a matter of course, for they declared motion itself

to be causeless (cf. Met. I. 4).

Page 64: Windelband, History of Philosophy, Vol. 1

44 The Greeks : Cosmological Period*. £Part L

which Empedocles and Anaxagoras, each in his way, had attempted,

was thus in turn abolished by the Atornists. They ascribed to the

particles of matter the capacity, not indeed of qualitative change

(dAXoiWis), but of independent motion (/cu/770-is in the narrower sense,

equivalent to 7repi<f>opa), and took up again in this sense the principle

of Milesian hylozoism.

10. In opposition to these pluralistic systems, Zeno, the friend

and disciple of Parmenides, sought to defend the Eleatic doctrine by

setting forth the contradictions in which the assumption of a plural-

ity of Beings is involved. As regards size, he pointed out, it fol-

lows that the totality of Being must be on the one hand infinitely

small, on the other hand infinitely great : infinitely small, because

the combination of any number whatever of parts, each of which is

to be infinitely small, never yields anything more than an infinitely

small sum

;

l infinitely great, on the contrary, because the bound-

ary which is to separate two parts must itself be an existent some-

thing, i.e. spatial magnitude, which again is itself separated from

the two parts by a boundary of which the same holds true, and so

on in infinitum. From the latter argument, which was called that

from dichotomy (the Ik Si^oi-o/Aias), Zeno reasoned also that as

regards number, what is must be unlimited, while, on the other hand,

this complete Being, not in process of becoming, is to be regarded

also as numerically limited [i.e. as complete]. And just as with the

assumption of the " many," so the position that empty space is real

is held to refute itself by a regress ad infinitum : if all that is is in

space, and thus space is itself an existing entity, then it must itself

be in a space, and this last likewise, etc. When the concept of the

infinite, to which the Atornists had given a new turn, became thus

prominent, all the enigmas involved in it for the contrasting points

of view of intellect and sense-perception became prominent also, and

Zeno used them to involve in a reductio ad absurdum the opponents

of the doctrine of the one, self-limited Being.

This dialectic, however, cut both ways, as was shown in the Ele-

atic School itself, by the fact that a cotemporary of Zeno, Melissus,

who shared his opinions, saw himself forced to declare that the

Being of Parmenides was as unlimited in space as in time. For as

Being can arise neither from other Being nor from Non-being, so

it can be limited neither by existing Being (for then there must be

a second Being), nor by a non-existent (for then this non-existent

must be) : a line of argument more consistent from a purely theo-

1 The argument can be directed only against Atomism, and applies to this

weakly.

Page 65: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 4.] Conceptions of Being : Pythagoreans. 45

retical point of view than the position of the master, which hadbeen influenced by determinations of worth.

11. The Pythagoreans took a mediating position in these ques-

tions : for this, as for their other doctrines, they were happily fitted

by their employment with mathematics, and by the manner in whichthey prosecuted this study. Its chief direction seems to have beenarithmetical ; even the geometrical knowledge ascribed to them (as

the well-known proposition named after Pythagoras) amounts to a

linear representation of simple relations between numbers (32 + 42

= 52, etc.). It was not, however, in the general relations of construc-

tions in space only that the Pythagoreans found numbers to be the

determining principles ; the same was found to be true also in such

phenomena of the corporeal world as they were chiefly engaged

with. Their theoretical investigations concerning music taught themthat harmony was based upon simple numerical relations of the

length of the strings (octave, third, fourth), and their knowledge

of astronomy, which was far advanced, led them to the view that

the harmony prevailing in the motions in the heavenly bodies had,

like the harmony in music, 1its ground in an order, in accordance

with which the various spheres of the universe moved about a com-

mon centre at intervals fixed by numbers. Suggestions so various

as these mentioned seem to have united to evoke in a man like

Philolaus the thought, that the permanent Being which philosophy

was seeking was to be found in numbers. In contrast with the

changing things of experience mathematical conceptions possess d*

regards their content the marks of a validity not subject to time

— they are eternal, without beginning, imperishable, unchangeable,

and even immovable ; and while they thus satisfy the Eleatic postu-

late for Being, they present, on the other hand, fixed relations,—that rhythmical order which Heraclitus had demanded. Thus, then,

the Pythagoreans found the abiding essense of the world in the

mathematical relations, and in particular in numbers,— a solution

of the problem more abstract than the Milesian, more capable of

being represented to perception or imagination than the Eleatic,

clearer than the Heraclitic, more difficult than those offered by

cotemporary mediating attempts.

The Pythagorean doctrine of numbers, as carried out by them, was

attached partly to the numerous observations they had made on the

arithmetical relations, partly to analogies which they discovered or

sometimes artificially introduced, between numerical and philosophi-

cal problems. The definite nature of each individual number and

1 Out of this analogy arose the fantastic idea of the harmony of the spheres.

Page 66: Windelband, History of Philosophy, Vol. 1

46 The Greeks : Cosmological Period. [Part I

the endlessness of the number series must indeed have at first sug-

gested that reality belongs as well to the limited as to the unlimited,

and by transferring this thought into the geometrical sphere the

Pythagoreans came to recognise, in addition to the elements as the

limited, a Reality as belonging also to space as the unlimited void.

They thought of the elements, however, as determined by the forms

of the simple solids : fire by the tetrahedron, earth by the cube,

air by the octahedron, water by the icosahedron, and a fifth material,

aether, which they added as the celestial element to the four terres-

trial elements assumed by Empedocles, by the dodecahedron. 1 In

these conceptions the prevailing idea was this : corporeality, or the

essential quality of bodies, consists in the mathematical limitation

of the unlimited, in the shaping out of space into forms. Mathemati-

cal forms are made the essence of physical reality.

The Pythagoreans further believed that in the antithesis between

the limited and the unlimited they recognised the antithesis found

in numbers between the odd and the even

;

2 and this antithesis was

again identified with that between the perfect and the imperfect,

the good and the bad,3 in this last case not without the influence of

old ideas connected with the religious faith of the oracles. Their

Weltanschauung becomes thus dualistic: over against the limited,

odd, perfect, and good stands the limitless, even, imperfect, and bad.

As, however, both principles are united in the number one,4 which

has the value of an even as well as of an odd number, so in the

world as a whole these antitheses are adjusted to form a harmony.

The world is harmony of numbers.

Some of the Pythagoreans,5 moreover, sought to trace out through

the various realms of experience that fundamental antithesis, in the

assumption of which all the school were agreed, and so a table of ten

pairs of opposites came into existence: viz. limited and unlimited—odd and even— one and many— right and left— male and female

— at rest and in motion— straight and curved— light and dark—

1 While the main line of the Pythagoreans thus followed Empedocles, a later,

Ecphantus, conceived of this limitation of space in the sense of Atomism.2 The reason presented for this, viz. that even numbers permit of bisection

to infinity (?), is indeed very questionable and artificial (Simpl. Phys. D. 105r

455, 20).3 Nor must we here overlook the factor which had already asserted itself with

Xenophanes and Parmenides, viz. that to the Greek the conception of measurewas one that had a high ethical worth ; so that the infinite, which derides all

measure, must to him appear imperfect, while the definite or limited (jreirepaa-

ßivop) was necessarily regarded as more valuable.* Arist. Met. I. 5, 986 a 19.5 Or men standing in close relations with Pythagoreanism, such as the physi-

cian Alcmseon, a perhaps somewhat older contemporary of Philolaus. Cf.

Arist. Met. I. 5, 986 a 22.

Page 67: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 5.] Conceptions of Cosmic Processes. 47

good and bad— square and oblong or with unequal sides. This is

evidently a collection put together without system, to fill out the

sacred number ten, but an attempt at an articulation may at least be

recognised.

In accordance, then, with this or a similar scheme the Pythagoreans

exerted themselves to make an order of things corresponding to the

system of numbers, by assigning the fundamental conceptions in

every department of knowledge to various numbers, and on the other

hand by adjudging to every individual number, but especially to those

from one to ten, determining significance in the various spheres of

reality. The fantastic nature of the symbolic interpretation into

which they fell in doing this must yet not cause us to overlook the

fact that the attempt was therewith made to recognise an abiding order

of things which could be grasped and expressed in conceptions, and to

find the ultimate ground of this order in mathematical relations.

Nor did it escape the notice of the Pythagoreans themselves,

notably of the later members of the school, that numbers could not

be called the principles (dpxa of things in the same way in which

the term is applied to the various "stuffs," or kinds of matter, to the

elements, etc., that things have not arisen out of them, but are

formed according to them; and perhaps they best and most effec-

tively express their thoughts when they say that all things are

copies or imitations of numbers. With this conception the world of

mathematical forms was thought as a higher, more original reality,

of which the empirical reality was held to be only a copy : to the

former belonged abiding Being ; the latter was the contrasted world

of Becoming and change.

§ 5. Conceptions of Cosmic Processes.1

E. Hardy, Der Begriff der Physis in griechischen Philosophie, I. Berlin, 1884.

As the fact of change— that is, the cosmic processes— furnished

the most immediate occasion for reflection upon the abiding Being,

so, on the other hand, the various conceptions of Being had

as their ultimate aim only to make the processes of Nature intel-

ligible. This task was indeed occasionally forgotten, or set aside,

in the development of the conceptions of Being, as by the Eleatics

;

but immediately afterward the further progress of thought proved

to be determined all the more by the renewed attention given to

1 [Geschehen. I have translated this word variously by "change," "occur-rence," "event," "taking place," "coming to pass," "becoming," etc. Thelast, which is ordinarily used for the Greek yiyvoßat seems hardly broad enough.

The German means any natural process or event.]

Page 68: Windelband, History of Philosophy, Vol. 1

48 The Grreeks : Cosmological Period. [Part I

Becoming and change, and by the need of so thinking Being that

Becoming and change could not only be reconciled with it, but also

be made intelligible by it. Hand in hand, then, with ideas of Being,

go those of Becoming, the two in constant relation to one another.

1. To the Ionians the living activity of the world was something

so much a matter of course that they never thought of asking for

a cause of it. Naive Hylozoism could have in view only the explana-

tion of a 'particular occurrence or cosmic process. Explanation,

however, consists in reducing what is striking— not a matter of

course or intelligible in itself— to such simpler forms of occur-

rence as seem to need no explanation, inasmuch as they are most

familiar to our perception. That things change their form, their

qualities, their working upon one another, seemed to the Mile-

sians to require explanation. They contented themselves in this

with conceiving these changes as condensation or rarefaction of the

cosmic matter. This latter process did not seem to them to need a

farther explanation, though Anaximenes at least did add, that these

changes in the state of aggregation were connected with changes in

temperature— condensation with cooling, rarefaction with growing

warm. This contrast gave rise to the arrangement of the states of

aggregation in a series corresponding to the degree of rarefaction

or condensation of the primitive matter

:

xviz. fire, air, water, earth

(or stone).

The Milesians used these ideas not only to explain individual

phenomena of Nature, particularly the meteorological processes so

important for a sea-faring people, but also to explain the develop-

ment of the present state of the world out of the prime matter.

Thus Thales conceived water as in part rarefying to form air and

fire, and in part condensing to form earth and stone ; Anaximenes,

starting from air, taught an analogous process of world-formation.

As a result of these views it was assumed that the earth— resting

on water, according to the first, on air, according to the second—occupied the centre of the sphere of air revolving about it, and this

sphere of air was yet again surrounded by a sphere of fire, which

either broke through or shone through in the stars.

In setting forth this process of world-origination, which was per-

haps still regarded by Thales and Anaximander as a process occur-

ring once for all, the Milesians attached themselves closely to the

cosmogonic poetry. 2 Not until later does the consideration seem to

1 Hence it is intelligible that there were also physicists (not known to us byname) who would regard the world-stuff as an intermediate stage between ail

and water, or between air and fire.2 Hence, also, the designation of the world-stuff as dpx^ (beginning).

Page 69: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 5.] Cosmic Processes : Anaximander, Heraclitus. 49

have gained prevalence, that if to change of form a change back to

the origiual form corresponds, and if, at the same time, matter is

to be regarded as not only eternal but eternally living, it is necessary

to assume a ceaseless process of world-formation and world-destruc-

tion, a countless number of successive worlds}

2. Although these essential constituents characterise also the

physical theories of Anaximander, he was led beyond them by his

metaphysical conception of the awapov. The infinite, self-moved

matter which was intended by this obscure conception was indeed,

as a whole, to have no definite properties. It was held, however, to

contain qualitative opposites within itself, and in its process of evolu-

tion to exclude them from itself, so that they became separate. 2

Anaximander remained then a Hylozoist in so far as he regarded

matter as self-moved; he had seen, however, that the differences

must be put into it if they were to come forth out of it on occasion

of its self-motion. If, then, as regards his doctrine of Being, he ap-

proached the later theory of a plurality of primitive substances, and

abandoned the doctrine that the primitive matter was changeable in

quality, he was yet entirely at one with the other Milesians as

regards his conception of the causelessness of the cosmic process, and

thought that by the union of the two opposites, the warm and the

cold, which he conceived as the first to come out from the äireipov, he

could explain water. This done, he could proceed with his cosmog-

ony along the oceanic path taken by Thales.

But besides these physical and metaphysical determinations, the

only fragment 3 preserved from him, giving his own words, repre-

sents the perishing of things as an expiation for injustice, and so

presents the first dim attempt to present the world-process as

ethical necessity, and to conceive of the shadows of transitoriness,

which rest even on the bright picture of Hellenic life, as retribution

for sin. However doubtful the particular interpretation of this

utterance, there is yet without doubt voiced in it the need of giving

to physical necessity the worth of an ethical order. Here Anaxi-

mander appears as a predecessor of Heraclitus.

3. The order of events which Heraclitus thought he could estab-

lish as the only constant amid the mutation of things, had two

essential marks, the harmony of opposites and the circuit completed by

1 This doctrine was supported, probably by Anaximander, certainly by

Anaximenes. It is repeated in Heraclitus and Empedocles.2 The decisive passages for this very controverted question (Ritter. Seydel,

Zeller) are Arist. Phys. I. 4, 187 a 20, and Simpl. Phys, ( DO M r 1">4, 14 (after

Theophrastus); also the continuation of the passage in the following note.

8 Simpl. Phys. CD.) Gr 24, 18. Cf. Th. Ziegler, Arch. f. Gesch. d. Philos.,

I. 16 ff.

Page 70: Windelband, History of Philosophy, Vol. 1

50 The Greeks : Cosmological Period. [Part 1.

matter in its successive changes in the universe. The observation that

everything in the world is in process of constant change was

exaggerated by Heraclitus to the claim that everything is con-

tinually changing into its opposite. The " other " was for him. eo

ipso the opposed. The "flux of things " became transformed in his

poetic rhetoric into a ceaseless strife of opposites, and this strife

(7rdA.e1u.os) he declared to be the father of things. All that seems to

be for a shorter or longer time is the product of opposed motions

and forces which in their operation maintain themselves in equilib-

rium. The universe is thus at every moment a unity divided in

itself and again re-united, a strife which finds its reconciliation, a

want that finds its satisfaction. The essence of the world is the

invisible harmony in which all differences and oppositions are

solved. The world is Becoming, and Becoming is unity of oppo-

sites.

These antitheses, according to the view of Heraclitus, present

themselves particularly in the two processes taking place in con-

trary directions, through which, on the one hand, fire becomes

changed into all things, and, on the other hand, all things change

back into fire. The same stages are passed through in both

processes: on the "way downward" fire passes over, by condensation,

into water and earth, on the "way upward" earth and water, by rare-

faction, pass over into fire ; and these two ways are alike. Change

and counter-change run on side by side, and the semblance of a per-

manent thing makes its appearance where for a time there is as

much counter-change upon the one way as there is change upon the

other. The fantastic forms in which Heraclitus put these views

envelop the essential thought of a sequence of changes taking place

in conformity to law, and of a continual compensation of these

changes. The world is produced from the fire in ever-repeated

rhythm and at fixed intervals of time, and then again flashes up in

fire, to arise from it anew, a Phoenix. 1

In this ceaseless transformation of all things nothing individual

persists, but only the order, in which the exchange between the

contrary movements is effected,— the law of change, which consti-

tutes the meaning and worth of the whole. If in the struggle be-

tween opposites it seems as though something new were constantly

arising, this new is at the same time always a perishing product.

The Becoming of Heraclitus produces no Being, as the Being of

Parmenides produces no Becoming.

1 In details his physical, and especially his astronomical, ideas are weak.Metaphysical inquiry is more important with him than explanatory investiga-

tion. He shares this with his opponent, Parmenides.

Page 71: Windelband, History of Philosophy, Vol. 1

chap. 1, § 5.] Cosmic Processes : Parmenides, Ernpedocles. 5t

4. In fact, the doctrine of Being held by the Eleatics excluded

with plurality and change, events or cosmic processes, also. Ac-

cording to their metaphysics an event or occurrence is incomprehen-

sible, it is impossible. This metaphysics tolerates no physics.

Parmenides denies to time, as to space, independent reality (aAAo

7rapeK roil eo'vros-) : for him there is only timeless Being with no dis-

tinctions. Although Parmenides added to the first part of his didac-

tic poem, which presents the doctrine of Being, a second part which

treats physical problems, this is yet done with the protest in advance

that he if here presenting not truth, but the " opinions of mortals."

At the basis of all these ordinary opinions lies the false presupposi-

tion, previously rejected, that in addition to Being there is still

another, Non-being. All becoming, all plurality and motion, rest on

the interaction of these opposites, which are then further designated

as light and darkness, warmth and cold. A Weltanschauung is then

portrayed in poetic imagery, in which fire shapes the dark empty

space into corporeal structures, a mode of representation which in

part reminds us of Heraclitus, and in part accords with the astro-

nomical teaching of the Pythagoreans. The all-ruling Fire-power

(Sat/xwv), as inexorable necessity (81/07), w^n the help of love (epax;)

forces together what is akin, working from the centre of the world

outward. Appropriation of the doctrines of others and polemic

against them appear in motley mixture, agreeably to ihe purpose of

the whole. Over this tissue thus interwoven hovers a poetic breath

of plastic formative power, but original research and clear concep-

tions are lacking.

5. Ideas more definite, and more usable for explaining the par-

ticular, are found among the successors, who transformed the Eleatic

conception of Being into the conceptions of element, homoiomerias,

and atom, expressly for this purpose. They all declare that by

occurrence or coming to be nothing else is to be understood than the

motion of unchangeable corporeal particles. Ernpedocles and Anax-

agoras seem still to have sought to connect with this the denial of

empty space,— a principle which they received from Parmenides.

They ascribed to their substances universal divisibility, and re-

garded parts as capable of displacement in such away that as these

parts mixed and reciprocally interpenetrated, all space should be

always filled out. The motion in the world consists, then, in this

1 The hypothetical exposition of how the world would have to he thoughl if.

in addition to Being, Non-being, plurality, and becoming were also regarded as

real, had, on the one hand, a polemic purpose; and on tbe other, it met the

want of his disciples, who probably demanded of the master an explanation of

his own of the empirical world.

Page 72: Windelband, History of Philosophy, Vol. 1

52 The Greeks : Cosmological Period. [Part I

displacement of the parts of matter, each of which is always crowd-

ing and displacing the other. Things at a distance from one another

cannot act upon one another, except as parts of the one flow out and

penetrate into the other. This action is the more possible in pro-

portion as the effluxes of the one body resemble in their spatial

form the pores of the other. So at least Empedocles taught, and

the assumption of an infinite divisibility of substances is attested in

the case of Anaxagoras also. Another picture of occurrence more

akin to the present way of thinking is that presented by Leucippus.

The atoms which impinge upon each other in empty space act upon

each other by pressure and impact, group themselves together, and

so form greater or smaller things or masses which are not separated

and destroyed until some impact or pressure of other masses comes

from without. All occurrence and coming to be consists in this

process in which atom-complexes are successively formed and

shattered.

The fundamental form of world-motion in all three systems, how-

ever, is that of the vortex, of circular rotation (Slvv)- According to

Empedocles it is brought about by the forces of love and hate acting

among the elements ; according to Anaxagoras it is begun by the

Reason-stuff acting according to ends, and then continues with

mechanical consistency ; according to Leucippus it is the result

always occurring from the collision of several atoms. The principle

of mechanism was with Empedocles still enveloped in myth, with

Anaxagoras it first made a half-successful attempt to break through

the covering, and was completely carried through only by Leucippus.

What hindered the first two from reaching this position was the

introduction of considerations of worth into their explanatory

theory. The one was for tracing the good and the evil back to cor-

responding powers of mind, which were, to be sure, not ascribed to

any being, but mythically hypostatised ; the other believed that he

could explain the order of the whole only from the assumption that

purposive, rationally considered impulse had originated the motions.

Yet both came so near the position of Leucippus as to demand a

teleological explanation for the beginning only of the vortex-motion;

the farther course of the motions, and thus every individual occur-

rence, they explained, as did Leucippus, purely mechanically, by the

pushing and crowding of the particles of matter after these are once

in motion in the manner determined. They proceeded so con-

sistently in this that they did not exclude from this mechanical

explanation even the origination and functions of organisms, amongwhich, moreover, plants are regarded as being as truly animate as are

animals. Anaxagoras is reproached for this by Plato and Aristotle,

Page 73: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 5.] Cosmic Processes : Anaxagoras, Leucippus. 53

and an expression of Empedocles has been handed down, 1 according

to which he taught that the animals had arisen here and there, with-

out any rule, in odd and grotesque forms, and that in the course of

time only those fitted for life maintained themselves. The principle

of the survival of the fittest, which plays so great a part in the

biology of to-day, i.e. in Darwinism, is here already clearly formu-

lated.

On the ground of these ideas, an interesting contrast discloses

itself in the case of the three investigators, as regards their atti-

tude toward cosmogonic theories. For Empedocles and for Leu-

cippus, namely, the process of world-formation and world-dissolu-

tion is a perpetual one ; for Anaxagoras, on the contrary, it is one

that takes place once for all. Between the first two there is again

the difference that Empedocles, like Heraclitus, teaches that the

world arises and perishes in periodic alternation ; while Atomism,

on the contrary, holds that a countless number of worlds come into

being and pass away. According to the principles of Empedocles,

to be more explicit, there are four different states of the elements

;

their complete intermixture, in which love alone rules, and hate is

excluded, he calls acfxupos2 (sphere) ; when hate penetrates, this

homogeneous world-sphere becomes separated into the individual

things, until the elements are completely parted from one another

;

and out of this separate condition love brings them again together,

until full union is again attained. Neither in the case of complete

mixture, nor in that of complete separation, are there individual

things ; in both cases the Eleatic acosmism makes its appearance.

A world of individual things in motion exists only where love and

hate struggle with one another in mingling and separating the

elements.

It is otherwise with Leucippus. Some of the atoms that dart

about irregularly in the universe strike together here and there.

From the various impulses to motion which the individual particles

bring with them, where such aggregations occur, there results,

according to mathematical necessity (dvay/o?), a whirling movementof the whole, which draws into itself neighbouring atoms and atom-

complexes, and sometimes even whole " worlds," and so gradually

1 Arist. Phys. II. 8, 198 b 29. Moreover, we find an expression alreadyattributed to Anaximander, which teaches a transformation of organisms byadaptation to changed conditions of life: Plut. Pine. V. 19, 1 (Doz. D. 430, 15).For man, also, the oldest thinkers claimed no other origin than that of growthout of the animal world: so Empedocles in Plut. Strom, fr. 2. (Dox. D. 579, 17).

2 Evidently not without suggestion from the Eleatic world-sphere, which this

absolute, fully adjusted mingling of all elements, taught by Empedocles, mucliresembles.

Page 74: Windelband, History of Philosophy, Vol. 1

54 The G-reeks : Cosmological Period. [Part L

extends. Meanwhile such a system in process of revolution is

differentiating itself, since, by the rotation, the finer, more movable

atoms are driven to the periphery, the more inert and massy are

gathered in the centre ; and so like finds its way to like, not by

inclination or love, but through their like conformity to the law of

pressure and impact. So there arise at various times and in differ-

ent places in the boundless universe, various worlds, each of which

continues in motion within itself, according to mechanical law, until

it perhaps is shattered in pieces by collision with another world, or

is drawn into the revolution of a greater. So, the Atomists main-

tained, the sun and moon were at one time worlds by themselves,

which subsequently fell into the greater vortex of which our earth

is the centre. How near in principle this whole conception is to

the natural science of to-day is obvious.

The teleological point of view taken by Anaxagoras excludes, on

the contrary, a plurality of worlds in time as well as a plurality of

worlds in space. The ordering mind, which introduces the pur-

posive motion of the elements, forms just this one world only, which

is the most perfect. 1 Anaxagoras, therefore, quite in the manner of

the cosmogonic poetry, describes how the beginning of the world

was preceded by a chaotic primitive condition, in which the ele-

ments were intermingled without order and without motion. Then

came the voCs, the "Reason-stuff" {Vernunftstoff), and set it into

ordered motion. This vortex-motion began at one point, the pole of

the celestial vault, and extended gradually throughout the entire

mass of matter, separating and dividing the elements, so that they

now perform their mighty revolution in a uniformly harmonious

manner. The teleological motive of the doctrine of Anaxagoras

is due essentially to his admiration of the order in the stellar

world, which, after it has performed the rotations started by the

vovs, moves on without disturbance always in the same track. There

is no ground for assuming that this teleological cosmology directed

attention to the adaptation to ends in living beings, or even to the

connected system of Nature as beneficent to man ; its gaze was fixed

on the beauty of the starry heavens ; and what is related of the

views of Anaxagoras on terrestrial things, on organisms, and on

man, keeps quite within the setting of the mechanical mode of

explanation in vogue among his contemporaries. What he said, too,

with regard to the presence of life on other heavenly bodies, might

just as well have come from the Atomists.

1 This motive, fully carried out, is found in Plato, Tim. 31, with unmistak-able reference to the opposition between Anaxagoras and the Atomists.

Page 75: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 5.] Cosmic Processes : Zeno, the Pythagoreans. 55

Accordingly, although Anaxagoras conceived of the wOs as also the principleof animation, and thought of the particles of this substance as mingled ingreater or lesser number with organic bodies, yet the central point in this con-ception is that of the authorship of the astronomical world-order. The otherside, the moment or factor of the cause of animate life, is much more energeti-cally emphasised in the transformation which a younger eclectic naturalphilosopher, Diogenes of Apollonia, undertook to effect in the conception ofAnaxagoras by connecting it with the hylozoistic principle of Anaximenes.He designated air as äpx*ö [first principle, primitive element], fitted it out,however, with the characteristics of the voDs, — omniscience and force actingaccording to ends, — named this "rational air" also irv*v/j.a [spirit], and foundthis formative principle in man and other organisms as well as in the universe.A rich physiological knowledge enabled him to carry through in detail thisthought as applied to the structure and functions of the human body. Withhim teleology became the dominant mode of apprehending also the organicworld.

His fragments have been collected by Schorn (Bonn, 1829) and Panzerbieter(Leips. 1830). Cf. K. Steinhart in Ersch und Grüber's Encyclopädie.

6. All these doctrines, however, presuppose the conception of

motion as one that is intelligible of itself and in need of no further

explanation. They thought they had explained qualitative changewhen they had pointed out as its true essence motion, whetherbetween the parts of a continuously connected matter, or in emptyspace. The opposition, therefore, which the Eleatic School brought

to bear upon all these doctrines was directed first of all against this

conception of motion, and Zeno showed that this could by no meansbe taken so simply, but was rather full of contradictions which inca-

pacitated it for serving as principle of explanation.

Among Zeno's famous proofs of the impossibility of motion, 1 the

weakest is that which proceeds from the relativity of the amount ofmotion, by showing that the movement of a wagon is variously esti-

mated if it is observed either from wagons also in motion but in

different directions and at varying rates of speed, or again from twowagons one of which is moving and one standing still. The three

other proofs, on the contrary, which made use of the analysis into

discrete parts, infinitely many and infinitely small, of the space

passed through by motion, and the time occupied by it, werestronger, and for a long time were not overcome. The first proof

was with reference to the impossibility of passing through a flxed

space. This was regarded as proved by the infinite divisibility of

the line, since the infinite number of points which must be attained

before reaching the goal permitted no beginning of motion. Thesame thought appears, somewhat varied, in the second argument,

which seeks to prove the impossibility of passing through a space

which has movable boundaries. The argument (known as that of

1 Arist. Phys. VI. 9, 239 b. 9. Cf . Ed. Wellmann, Zenon's Beweise gegen dieBewegung und ihre Widerlegungen (Frankfurt a. 0. 1870).

Page 76: Windelband, History of Philosophy, Vol. 1

56 The G-reeks : Cosmological Period. [Part I

Achilles and the tortoise) is, that since the pursuer in every inter

val or subdivision of time must first reach the point from which the

pursued simultaneously starts, it follows that the latter will always

be in advance, though by an interval which becomes constantly

smaller and approaches a minimum. The third argument has refer-

ence to the infinitely small extent of the motion performed in any

instant. According to this argument, called u the resting arrow" the

moved body is in every instant in some one point of its track ; its

movement in this instant is then equal to zero ; but from ever so

many zeros no real magnitude arises.

Together with the above-mentioned difficulties (airopiai) with

regard to space and plurality, these argumentations of Zeno set

forth an extremely skilfully projected system of refuting the

mechanical theories, especially Atomism, — a refutation which was

intended to serve at the same time as indirect proof of the correct-

ness of the Eleatic conception of Being.

7. The number-theory of the Pythagoreans, too, was determined by

Eleatic conceptions in so far as its procedure was, in the main, to

demonstrate mathematical forms to be the fundamental relations

of reality. When, however, they termed the actual world of reality

an imitation of the mathematical forms, they thereby ascribed a sort

of reality, even though of a derivative and secondary character, to

individual things, and to what takes place among them. They were

also the less inclined to withdraw from answering cosmological and

physical questions as they were able to bring to philosophy the

brilliant results of their astronomical investigation. They had come

to a knowledge of the spherical form of the earth and of the heav-

enly bodies ; they were aware also that the change of day and night

depends upon a movement of the earth itself. At first, indeed, they

thought of this movement as a circuit performed about a central fire

to which the earth presented always the same side, a side unknownto us. 1 On the other hand, they assumed that about this same cen-

tral fire there moved in concentric circles, outside the earth's track,

successively the moon, the sun, the planets, and finally the heaven

containing the fixed stars. They brought into this system, however,

in a way, the metaphysical dualism which they had maintained be-

tween the perfect and the imperfect, inasmuch as they regarded the

1 Already in Plato's time the hypothesis of the central fire was given up bythe younger Pythagoreans, Ecphantus, Hicetus of Syracuse (and with it thatof the " counter-earth," which had hitherto been assumed as placed between thecentral fire and the earth, invented merely to fill out the number ten), andinstead the earth was located in the centre of the universe and provided with arotation on its axis. With this latter assumption that of a resting position of

the heaven of the fixed stars was connected.

Page 77: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 6.] Conceptions of Cognition. 57

heaven of the stars, on account of the sublime uniformity of its

motions, as the realm of perfection; the world "beneath the moon,"

on the contrary, on account of the unrest of its changing formations

and motions, they regarded as that of imperfection.

This way of looking at things runs parallel to that of Anaxagoras,

and leads, though in another way, to the interweaving and complica-

tion of theory with considerations of worth [ethical or aesthetic

values]. It ivas in connection ivith astronomical insight that the

thought of an order of Nature in conformity to law dawned as clear

knowledge upon the Grecian mind. Anaxagoras reasons from this

to an ordering principle. Pythagoreanism finds in the heavens the

divine rest of unchangeableness (Sichgleichbleibens) which it misses

upon the earth. Here we have a meeting of the ancient religious

ideas and the very different result yielded thus far by the scientific

work of the Greeks. This latter, seeking a Permanent in the muta-

tion of occurrence, found such a permanence only in the great, simple

relations, in the revolution of the stars, which abides ever the same.

In the terrestrial world, with its whole change of manifold, con-

stantly intersecting motions, this uniformity remained still hidden

from Greek science : she regarded this terrestrial world rather as a

domain of the imperfect, the lower, which wants the sure order of

that other world. In a certain sense this may be looked upon as

the ultimate result of the first period, a result which had a determin-

ing influence for after time.

What the attitude of the Pythagoreans was to the question concerning a peri-

odic change of origination and annihilation of the world is uncertain. A plurality

of co-existing worlds is excluded in their system. In their theory of world-for-

mation and in their particular physical doctrines they concede so prominent aplace to fire that they come very near to Heraclitus. Aristotle even places oneof the contemporaries of Philolaus, Hippasus of Metapontum, in immediate con-nection with Heraclitus (Met. I. 3).

Their assumption of sether as a fifth element out of which the spherical shells

of the heavens were formed, in addition to the four elements of Empedocles, is

doubtless connected with the separation which they made between heaven andearth. It is not less difficult to decide whether they derived the elements froma common ground, and if so, how : according to many passages it would seem asif they had spoken of a progressive " attraction," i.e. in this case (cf. above, p.

46), mathematical shaping out or forming of empty space by the '4v (one), the

original number, which is exalted above limitation and the unlimited. Yet it

seems, too, that in regard to these questions various views were held within theschool side by side.

§ 6. The Conceptions of Cognition.

M. Schneidewin, Ueber die Keime erkenntnisstheoretischer und ethischer Phi-

losopheme beiden vorsokratischen Denkern, Philos. Monatshefte, II. (1869), pp.257, 345, 429.

B. Münz, Dip Keime der Erkenntnisstheorie in der vorsophistischoi Periodeder griechischen Philosophie. Vienna, 1880.

Page 78: Windelband, History of Philosophy, Vol. 1

58 The Gfreeks : Cosmological Period. [Part I

The question, what things really are, or what is the intrinsic

nature of things, which is already contained in the Milesian con-

ception of the apxV) presupposes that the current, original and na'ive

mode of thinking of the world has been shaken, although this pre-

supposition has not come to clear recognition in consciousness. Thequestion proves that reflective thought is no longer satisfied with

the ideas which it finds current, and that it seeks truth behind or

above them. Those ideas are given, however, through sense-per-

ception and through the involuntary elaboration of this in thought,

— an elaboration that has been transmitted from generation to

generation, until it has became consolidated and fixed and embodied

in language, and so forms a part of the thinker's data. When the

individual with his reflection transcends these ideas so given— and

it is in this that philosophical activity ultimately consists— he does

it on the ground of logical needs which assert themselves as he re-

flects on the given. His philosophising, then, even though he takes

no account of this fact, grows out of discrepancies between his expe-

rience and his thought— out of the inadequacy exhibited by what

is presented to his perception or imagination, when set over against

the demands and presuppositions of his understanding. Howeverunconscious of this its inner ground na'ive philosophising may be

at the outset, attention cannot fail to be turned in time to the diver-

sity in the sources of the conflicting ideas within.

1. The first observations, therefore, which the Grecian philosophers

made on human knowledge concern this contrast between experience

and reflection. The farther the explanatory theories of science

became separated from the way of looking at things which belongs

to daily life, the clearer it became to their authors that those

theories sprang from another source than that of the customary

opinions. To be sure they have not as yet much to say on this

point. They set opinion (So£a) over against truth, and this often

means only that their own doctrines are true and the opinions of

others false. So much only is certain to them, that they owe their

own views to reflection, while the mass of mankind— concerning

whose intellectual activity it is just, the older philosophers,

Heraclitus, Parmenides, Empedocles, who express themselves in

an extremely depreciatory manner— persist in the illusion of the

senses. Only through thinking (<j>povelv, votlv, Aoyos), then, is the

truth found ; the senses, if alone, give fraud and a lie.l So strong

has reflection become in itself that it not only proceeds to con-

sequences which to the common thinking have become absolutely

1 Heracl. Frag. (Schust.) 11, 123 ; Pannen. Frag. (Karsten) 54 ff.

Page 79: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 6.] Conceptions of Cognition: Heraclitus, Parmenides. 59

paradoxical, but also maintains expressly that it is itself the sole

source of truth as opposed to opinions.

This, to be sure, works oddly when we notice that completely

opposite illustrations of this same assertion are given by Heracli-

tus and Parmenides in close succession. The former finds the

deceit caused by the senses, and the error of the multitude, to consist

in the illusory appearance of the Being of permanent things, which

is presented to men by sense-perception ; the Eleatic, on the contrary,

is zealous against the senses, because they would fain persuade us

that there are in truth motion and change, becoming and arising,

plurality and variety. Precisely this double form in which this

same claim is put forward shows that it is not the result of an

investigation, but the expression of a demand made on other

grounds.

Moreover, this proposition fits very differently into the general

theories of the two great metaphysicians. The flux of all things,

with its restless change of individual phenomena, as taught by

Heraclitus, makes it easy to comprehend also the possibility of the

emergence of false ideas, and the seeming of permanence and Being

had besides a special explanation in the counter-course or opposi-

tion (evavTtoTpo7rta) of the two " ways," for this causes the illusion of

permanence or Being to arise where there is just as much change in

one direction as in the other [i.e. from primitive fire into things and

vice versag . On the contrary, it is quite impossible to see where the

seat of illusion and error was to be sought in the one world-sphere

of Parmenides, everywhere the same, which was held to be at the

same time the one, true world-thought. The search could be only

among individual things and their changing activities, which were

themselves declared to be illusion, non-existent. Nevertheless

there is no support to be found in the literature preserved, for

supposing that this so simple a thought * which would have ove^

thrown the entire Eleatic system, ever occurred to the investigators

of that time. In any case, the Eleatics contented themselves with

the assertion that all particular existence and all change were decep-

tion and illusion of the senses.

The same na'ive denial of that which they coidd not explain seems to

have been employed also by the successors of the Eleatics in the

matter of the qualitative attributes of individual things. Emped-

odes at least maintained that all things were mixtures of the ele-

ments. The task that logically grew out of this was to show howthe other qualities arise from the mixture of the properties of the

1 First carried out in Plato, Sophist, 237 A.

Page 80: Windelband, History of Philosophy, Vol. 1

60 The Greeks : Cosmological Period. [Part I.

elements. But this he did not perform; so far as our knowledge,

extends, he did not at all set himself this task; he probably re^

garded these particular qualities as not being (objectively), and as

a deception of the senses, just as all qualities whatever were such

in the view of Parmenides. And so the oldest view of the Ato-

mists, as supported by Leucippus, may well have gone just to this

point, maintaining that in individual things only the form, arrange-

ment, situation, and motion of the constituent atoms were real, and

that the other properties were a deceitful product of the senses,

which here, too, found no further explanation. 1

These difficulties were perhaps jointly influential in the mind of

Anaxagoras when he regarded all qualities as original, and not as

having become what they are, and accordingly postulated countless

elements. But for him arose the opposite difficulty of showing howit could come about, if all was regarded as contained in all, every

quality in every thing, that only some of these qualities seemed to

be present in individual things. He explained this in part from the

consideration that many of the constituent parts are imperceptible

because of their minuteness ; hence it is only by thought that wecan learn the true qualities of things. 2 Besides this, however, he

seems to have followed up the thought, found already in Anaximan-

der's idea of the awupov, that a complete mingling of definite quali-

ties yields something indefinite. So, at least, he described the

primitive mixture of all substances which preceded the formation

of the world as completely devoid of quality,3 and a similar thought

seems to have permitted him to regard the four elements of Emped-

ocles not as primitive substances, but rather as already mixtures.4

The rationalism common to the pre-Sophistic thinkers assumes,

among the Pythagoreans, the particular form of affirming that

knowledge consists in mathematical thought. This, though in itself

a narrowing, is yet, on the other hand, a great step in advance, in-

asmuch as there is here given for the first time a positive definition

of " thought " as contrasted with " perception." Only through

number, taught Philolaus, 5is the essential nature of things to be

known ; that is, it is when the definite mathematical relations lying

at their basis are recognised that things are properly conceived or

1 It is extremely improbable that the solution of the problem through thesubjectivity of the sense-qualities, which is found in Democritus, was presentedalready by Leucippus, and therefore before Protagoras, who is universally

regarded as the founder of this theory.2 Sext. Emp. Adv. Math. VII. 90 f.

3 Frag. (Schorn) 4. From this passage the true light may, perhaps, be thrownupon the sense in which Anaximander designates the äireipov as dopia-rov.

* Arist. De Gen. et Corr. I. 1, 314 a 24.8 Frag. (Mull.) 13.

Page 81: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 6.] Conceptions of Cognition : Philolaus, Zeno. 61

understood. This had been the experience of the Pythagoreans in

music and in astronomy, and this was the object of their desire andeffort in all other fields. When, however, they ultimately came to

the result that this requirement could be completely met only in

the knowledge of the perfect world of the stars, they concludedfrom this that science (o-o^ta) relates only to the realm of order andperfection, that is, to heaven, and that in the realm of the imper-

fect, of change not subject to order, i.e. on earth, only practical

ability (ipcTrj) is of avail. 1

Another positive characteristic of the "thinking" which the

earlier investigators had set over against "perceiving," withoutcloser specification, appears obscurely in the reasonings of Zeno,

viz. conformity to logical laws. At the basis of all his attacks

against plurality and motion lie the principle of contradiction andthe presupposition that that can not be actual of which the samething must be affirmed and also denied. This principle and presup-

position were applied with clearness and certainty, though not ab-

stractly expressed. The Eleatic theory of the world, so highly

paradoxical, forced its supporters to enter into polemic more thandid others, and the accounts as to Zeno's treatise, which, as it seems,

was also logically well arranged and divided, offer a notable evi-

dence of the developed technique of refutation to which the school

attained in consequence. To be sure, this formal training whichprevailed in Eleatic circles does not seem to have led as yet to the

abstract statement of logical laws.

2. The setting over against each other of " thinking " and " per-

ceiving " arose, then, from an estimation of their relative epistemo-

logical value (erkenntnisstheoretischen Werthbestimmung) [i.e. fromthe postulate that one of these two forms of mental activity is

worth more epistemologically for attaining truth]. In decided

contradiction with this, however, stand the psychological principles

with which these same investigators sought to apprehend the origin

and process of knowing. For although their thinking was directed

first and chiefly toward the outer world, man's mental activity cameunder their attention in so far as they were obliged to see in this

activity one of the formations, or transformations, or products of

motion, of the universe. The mind or soul and its action are thenat this time considered scientifically only in connection tvith the entire

course of the universe, whose product they are as truly as are all

other things ; and since among the men of this period the general

principles of explanation are everywhere as yet conceived corpore-

i Stob. Eel. I. 488.

Page 82: Windelband, History of Philosophy, Vol. 1

62 The Greeks : Cosmological Period. [Part L

ally it follows that we meet also a thorough-going materialistic

psychology}

Now mind or soul is in the first place moving force. Thales

ascribed such a soul to magnets, and declared that the whole world

was full of souls. The essential nature of individual souls was

therefore sought at first in that which had been recognised as the

moving principle in the whole. Anaximenes found it in air,

Heraclitus and likewise Parmenides (in his hypothetical physics)

in fire, Leucippus in the fiery atoms, 2 and Anaxagoras in the world-

moving, rational substance, the voBs. Where, as in the system of

Empedocles, a corporeal moving principle was lacking, the mixed

substance which streams through the living body, the blood, was

regarded as soul. Diogenes of Apollonia found the essence of the

soul in the air mixed with the blood.3 With the Pythagoreans, too,

the individual soul could not be considered as the same with the h(One) which they conceived as moving principle of the world, nor

regarded as a part of it ; instead, they taught that the soul was a

number, and made this very vague statement more definite by say-

ing that it was a harmony, — an expression which we can only

interpret 4 as meaning a harmony of the body; that is, the living,

harmonious activity of its parts.

If now to this moving force, which leaves the body in death, were

ascribed at the same time those properties which we to-day designate

as " psychical," we find a clear characterisation of the specifically

theoretical interest by which this oldest science was filled, in the

fact that among these attributes it is that of ideation, of " knowing,"

which is almost exclusively the object of attention.5 Of feelings

and volitions there is scarcely incidental mention.6 But as the

1 Besides those characterisations of the soul, which resulted from their gen-eral scientific theory, we find in the tradition in case of several of these men(Heraclitus, Parmenides, Empedocles, and the Pythagoreans) still other doc-trines which are not only not connected with the former, but are even in con-tradiction to them. A conception of the body as prison of the soul (<r<2/xa =<rrjfxa), personal immortality, recompense after death, transmigration of souls,— all these are ideas which the philosophers took from their relations to the

mysteries and retained in their priestly teaching, however little they accordedwith their scientific teachings. Such expressions are not treated above.

2 In like manner, some of the Pythagoreans declared the motes which thesunlight discloses in the air to be souls.

8 Since, with reference to this, he recognised the distinction between venous andarterial blood, he meant by his wvedßa what the chemistry of to-day calls oxygen.

4 Ace. to Plato, Phcedo, 85 ff., where the view is rejected as materialistic.5 The vovs of Anaxagoras is only knowing ; air with Diogenes of Apollonia is

a great, powerful, eternal, intelligent body. Being with Parmenides is at the

same time voetv, etc. Only ^iXott/s and velnos with Empedocles are mythically

hypostasised impulses, and these, too, have nothing to do with his psychological

views.6 With this is connected the fact that in general we cannot once speak of

Page 83: Windelband, History of Philosophy, Vol. 1

Chap. I, § 6.] Cpnceptions of Cognition: Heraclitus, Anaxagoras. 63

individual soul in so far as it is moving force was held to be a part

of the force which moves the entire universe, so also the " knowing "

of the individual could be conceived only as a part of the knowing

activity of the world. 1 This S clearest in the systems of Heraclitus

and Anaxagoras ; each individual has so much knowledge as there

is contained in him of the general World-reason,— fire with

Heraclitus,- the vov<s with Anaxagoras. In the case of Leucippus

and of Diogenes of Apollonia the ideas are similar.

This physical conception, which with Anaxagoras especially is

purely quantitative, was given a turn by Heraclitus, in which the

epistemological postulate again forces its way to the front, and

asserts itself in the interest of a deeper insight and a profounder

view. The World-reason in which the individual participates in his

knowledge is everywhere the same; the Aoyos of Heraclitus 3 and

the i/oSs of Anaxagoras, as homogenous Reason, are distributed

through the whole universe as moving force. Knowing, then, is

that which is common to all. It is therefore the law and order to

which every one has to unite himself. In dreams, in personal opin-

ion, each one has his own world; knowing is common (£wov) to

all. By means of this characteristic, viz. that of universally valid

law, the conception of knowing acquires a normative significance*

and subjection to the common, to the law, appears as a duty

in the intellectual realm as well as in the political, ethical, and

reliarious.5

attempts at ethical investigation in this period. For single moralising reflections

or admonitions cannot be regarded as beginnings of ethics. On the only excep-tion cf. below, note 5.

1 The expression " World-soul " was first used by Plato, or at the earliest byPhnolaus (in the fragment which has certainly been much questioned just for

this reason, Mull. 21). The idea is certainly present in Anaximenes, Heraclitus,

Anaxagoras, and perhaps also among the Pythagoreans.2 Hence the paradoxical expression, the dryest soul is the wisest, and the

warning to guard the soul from the wet (intoxication).3 Cf., for this and the following, M. Heinze, Die Lehre vom Logos in der

griechischen Philosophie (Oldenburg, 1872).4 Frag. (Schust.) 123.5 This is the only conception in the development of pre-Sophistic thought, in

the case of which we can speak of an attempt to propound a scientific principle

of ethics. If Heraclitus had in mind a universal expression for all moral dutiesin speaking of this subordination to law, or at least hit upon such, he attachedit at once to the fundamental thoughts of his metaphysics, which declared this

law to be the abiding essence of the world. Yet attention has above (§ 4) beencalled to the iact that in the conception of the world-order which hovered beforehim, he did not as yet separate consciously the different motives (especially thephysical from the ethical), and so ethical investigation does not as yet workitself clear from the physical to an independent position. The same is true ofthe Pythagoreans, who expressed the conception of order by the term " harmony "

(which also might be adopted from Heraclitus), and therefore designated virtueas "harmony." To be sure, they used the term "harmony" for the soul, for

health, and for many other things.

Page 84: Windelband, History of Philosophy, Vol. 1

64 The Greeks : Cosmological Period. [Part 1

3. If now we ask how under these assumptions the fact was

explained that "knowledge" comes into the individual man, i.e. into

his body, we find that the only answer offered by Heraclitus and

the whole company of his successors is, " through the door of the

senses." When a man is awake, the World-reason streams into his

body through the opened senses (sight and hearing are of course

chiefly noticed 1

), and, therefore, he knows. This comes about, to

be sure, only if there is besides, in the man himself, so much reason

or soul that the motion coming from without is met by an inner

motion

;

2 but upon this interaction, effected through the senses»

between the outer and the inner reason knowledge rests.

A psychological distinction, then, between perceiving and think-

ing, which, as regards their respective epistemological values, are so

abruptly opposed, Heraclitus does not know how to state. Par-

menides,3 however, was just as little in a position to make such a

distinction. 4. Rather, he expressed more sharply still the dependence

upon bodily relations in which the thinking of the individual man is

involved, when he said that every one so thought as the conditions

constituted by the mixture of substances in the members of the body

permitted, and when he found in this a confirmation of his general

thought of the identity of corporeality and thinking in general.*

Still more express is the testimony 6 that Empedocles declared

thinking and perceiving to be the same, that he thought change in

thinking as dependent upon change of the body, and that he

regarded the constitution of the blood as of decisive importance

for the intellectual capacity of the man.

These two last-named thinkers did not hesitate, moreover, to maketheir conception more plain to the imagination by means of physio-

logical hypotheses. Parmenides taught in his hypothetical physics

1 Also smell (Empedocles) and taste (Anaxagoras). Only the Atomists, andin particular Democritus, seem to have given value to the sense of touch.

2 Arist. De An. I. 2, 405 a 27.3 Theophr. De Sens. 3 f.

4 So, too, it is reported (Theophr. De Sens. 25) of Alcmseon, the Pythago-reanising physician, that he declared thought or consciousness (ö'ti /xdvos ZwIijo-l)

to be the characteristic which distinguishes man from the other animals. But,

a more precise determination is lacking here also unless, in accordance with theexpression, we think of something similar to the Aristotelian koivov alad-qTrjpiov.

With this would agree the circumstance that the first attempts to localise theparticular psychical activities in particular parts of the body seem to have beenmade in the circles of the Pythagoreans and of the physicians who stood in nearrelations to them ; localising, e.g., thought in the brain, perception in the indi

vidual organs and in the heart, and the emotions also in the latter organ. Fromthem Diogenes of Apollonia, and after him Democritus, seem to have takenthese beginnings of a physiological psychology.

5 Frag. (Karst.) vv. 146-149.s Arist. De An. I. 2, 404 b 7 ; III. 3, 427 a 21; Met. III. 5, 1009 b 17;

Theophr. De Sens. 10 f.

Page 85: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 6.] Conceptions of Cognition : Parmenides, Empedocles. 65

that like is always perceived by like, warmth without by the warmthin man, the cold without by the cold even in the dead body. Emped-ocles, with the aid of his theory of effluxes and pores, carried out

the thought that every element in our body perceives the same ele-

ment in the outer world, so as to teach that each organ is accessible

to the impress of those substances only whose effluxes fit into its

pores ; i.e. he derived the specific energy of the sense organs from

relations of similarity between their outer form and their objects,

and carried this out for sight, hearing, and smell, with observations

which in part are very acute. 1

This view, that like is apprehended by like, was opposed byAnaxagoras,—on what ground it is not certain. 2 He taught that

perception is only of opposite by opposite, warmth without by the

cold in man, etc.3 At all events, his doctrine also is a proof that

these metaphysical rationalists maintained all of them in their

psychology a crass sensationalism.

1 Theophr. De Sens. 7.

2 Perhaps we have here a remembrance of Heraclitus, who also explainedperception from the ivavTioTpo-rrla, — motion against motion, — and with whomopposition was the principle of all motion.

3 Theophr. Be Sens. 27 ff. It is interesting that Anaxagoras inferred fromthis that every perception is joined with pain (\6irr)).

Page 86: Windelband, History of Philosophy, Vol. 1

CHAPTER IL

THE ANTHROPOLOGICAL PERIOD.

G. Grote, History of Greece, VIII. (London, 1850), pp. 474-544.

C. F. Hermann, Geschichte und System der platonischen Philosophie, I. (Heidel-

berg, 1839), pp. 179-231.

Blass, Die attische Beredsamkeit von Gorgias bis zu Lysias. Leips. 1868.

H. Köchly, Sokrates und seinVolk, 1855, in "Akad. Vorträgen und Reden," L

(Zürich, 1859), pp. 219 ff.

H. Siebeck, lieber Sokrates11

Verhältniss zur Sophistik, in "Untersuchungen

zur Philosophie der Griechen," 1873, 2 Aufl. (Freiburg i. B. 1888).

W. Windelband, Sokrates in "Prseludien" (Freiburg i. B. 1884), pp. 54 ff.

[H. Jackson, Art. Sophists, in Enc. Brit.]

The farther development of Greek science was determined by the

circumstance that in the powerful, universal upward movement of

the mental and spiritual life which the nation achieved after the

victorious result of the Persian wars, science was torn away from

the restraints of close schools in which it had been quietly pursued,

and brought out upon the stage of publicity, where all was in vehe-

ment agitation.

The circles in which scientific research was fostered had widened

from generation to generation, and the doctrines which at first had

been presented in smaller societies and spread abroad in writings

that were hard to understand, had begun to filter through into the

general consciousness. The poets, as Euripides and Epicharmus,

began already to translate into their language scientific conceptions

and views ; the knowledge gained by investigation of Nature hadalready been made practically effective, as by Hippodamus in his

architecture. Even medicine, which had formerly been only an art

practised according to traditions, became so permeated with the

general conceptions of natural philosophy, and with the special doc-

trines, information, and hypotheses of physiological research whichin the course of time had occupied an ever-broader space in the

systems of science, that it became encumbered with an excessive

Page 87: Windelband, History of Philosophy, Vol. 1

Chap. 2.] The Anthropological Period. 67

growth of etiological theories, 1 and first found in Hippocrates the

reformer who reduced this tendency to its proper measure and gave

back to the physician's art its old character in contrast to scientific

doctrine. 2

Moreover, the Greek nation, matured by the stern experience

which had been its lot within and without, had entered upon the

age of manhood. It had lost its naive faith in old tradition, and

had learned the value of knowledge and ability for practical life.

Of science, which up to this time had followed in quiet the pure

impulse of investigation— the noble curiosity which seeks knowledge

for its own sake— the state now demanded light on the questions

which disturbed it, counsel and help in the doubt into which the

luxuriance of its own development in culture had plunged it. In

the feverish emulation of intellectual forces which this greatest

period in the world's history brought with it, the thought everywhere

gained recognition that in every walk in life the man of knowledge

is the most capable, the most useful, and the most successful. In

every department of practical activity, the fruitful innovation of

independent reflection, of individual judgment, took the place of the

old life controlled by custom. The mass of the people was seized with

the burning desire to make the results of science its own. It was espe-

cially true, however, that at this time family tradition, habituation,

personal excellence of character and address were no longer suffi-

cient, as formerly, for the man who wished to play a political part.

The variety of transactions and the attendant difficulties, as well as

the intellectual status of those with whom and upon whom he wouldwork, made a theoretical schooling for the political career indispen-

sable. Nowhere was this movement so powerful as in Athens, then

the capital of Greece, and here also these desires found their fullest

satisfaction.

For the supply followed the demand. The men of science, the

Sophists (cro(£io-Tcu) , stepped forth out of the schools into public life,

and taught the people what they themselves had learned or discov-

ered. They did this, indeed, partly out of the noble impulse to

teach their fellow-citizens,3 but it was none the less true that this

teaching became their business. From all parts of Greece men of

the different schools flocked toward Athens to expound their doc-

1 This innovation in medicine began among the physicians who stood in nearrelation to Pythagoreanism, especially with Alcmseon. As a literary instanceof it, the writing which goes falsely under the name of Hippocrates, -n-epi Siairrjs,

serves. Cf. II. Siebeck, Gesch. d. Psych. I. 1, 94 if.

2 Cf. principally his writings 7repi äpyai-qs h]TptK7js and Trepl dtairrjs öi-tuv.3 Cf. Protagoras in Plato, Prot. 310 d.

Page 88: Windelband, History of Philosophy, Vol. 1

68 The Philosoph// of the (rreeks. [Part I.

trines, and from so expounding them in the capital as well as in

the smaller cities, to gain honour and wealth.

In this way it happened that in a short time not only the social

position of science, but its own inner nature, its tendency and the

questions for its solution, were fundamentally changed. It became

a social power, a determining factor in political life, as in the case

of Pericles ; but just by this means it came into a state of dependence

upon the demands of practical, and in particular, of political life.

These demands showed themselves principally in the facts that

the democratic polity demanded of politicians first of all the capac-

ity for public speaking, and that in consequence the instruction

of the Sophists was especially sought as a preparation for public

life, and converged more and more upon this object. Men of science

became teachers of eloquence.

As such, however, they lost sight of the goal of nature-knowledge,

the vision of which had formerly hovered before the eyes of science.

At the most they presented transmitted doctrines in the most grace-

ful and pleasing form possible. But their own investigations, if

they were not confined to a formal routine, were necessarily directed

toward man's thinking and loilling, — the activities which public

speaking was designed to determine and control,— toward the

manner in which ideas and volitions arise, and the way in which

they contend with one another and maintain their mutual rights.

In this way Greek science took an essentially anthropological or

subjective direction, studying the inner activities of man, his

ideation and volition, and at the same time lost its purely theoretical

-character and acquired a preponderantly practical significance}

But while the activity of the Sophists found itself brought face

to face with the manifold character of human thought and will,

while the teachers of eloquence were presenting the art of persua-

sion and pursuing the path upon which every opinion could be

helped to victory, every purpose to its achievement, the question

rose before them whether above and beyond these individual opin-

ions and purposes which each one feels within himself as a necessity

and can defend against others, there is anything whatever that

is right and true in itself. The question whether there is anything

universally valid, is the problem of the anthropological period of

Greek philosophy, or of the Greek Enlightenment.

For it is likewise the problem of the time, — of a time in which

religious faith and the old morality were wavering, a time when the

1 Cicero's well-known expression (Tusc. V. 4, 10) with regard to Socrates

holds good for the entire philosophy of this period.

Page 89: Windelband, History of Philosophy, Vol. 1

Chap. 2.] The Anthropological Period. 69

respect which authority had commanded sank more and more, and

all tended towards an anarchy of individuals who had become self-

governing. Very soon this internal disintegration of the Greek

spirit became clearly evident in the disorders of the Peloponnesian

war, and with the fall of Athenian supremacy the flower of Grecian

culture withered.

The dangers of this condition were at first decidedly increased by

philosophy. For while the Sophists were perfecting the scientific

development of the formal art of presentation, verification, and refu-

tation which they had to teach, they indeed created with this rheto-

ric, on the one hand, the beginnings of an independent psychology,

and raised this branch of investigation from the inferior position

which it had taken in the cosmological systems to the importance of

a fundamental science, and developed, on the other hand, the prelim-

inaries for a systematic consideration of the logical and ethical norms.

But as they considered what they practised and taught,— viz. the

skill to carry through any proposition whatever, 1— the relativity of

human ideas and purposes presented itself to their consciousness so

clearly and with such overwhelming force that they disowned in,

quiry as to the existence of a universally valid truth in the theoreti-

cal, as well as in the practical sphere, and so fell into a scepticism

which at first was a genuine scientific theory, but soon became a

frivolous play. With their self-complacent, pettifogging advocacy,

the Sophists made themselves the mouth-piece of all the unbridled

tendencies which were undermining the order of public life.

The intellectual head of the Sophists was Protagoras; at least, he

was the only one who was the author of any conceptions philosophi-

cally fruitful and significant. Contrasted with him, Gorgias, who is

usually placed at his side, appears only as a rhetorician who occa-

sionally attempted the domain of philosophy and surpassed the

artifices of the Eleatic dialectic. Hippias and Prodicus are only to

be mentioned, the one as the type of a popularising polyhistor, and

the other as an example of superficial moralising.

To the disordered activity and lack of conviction of the younger

Sophists, Socrates opposed faith in reason and a conviction of the

existence of a universally valid truth. This conviction was with

him of an essentially practical sort; it was his moral disposition, but

it led him to an investigation of knowledge, which he anew set over

against opinions, and whose essence he found in conceptional thought.

Socrates and the Sophists stand, accordingly, on the ground of

1 Cf. the well-known Tbv tJttw X670f KpdTTa irotelv, Aristoph. ,\n!i 112 ff.,

893 ff. ; Arist. Rhet. II. 24, 1402 a 23. .

Page 90: Windelband, History of Philosophy, Vol. 1

70 The Philosophy of the Greeks. [Pakt I.

the same common consciousness of the time, and discuss the same

problems ; but where the Sophists with their skill and learning re-

main caught in the confusion of the opinions of the day and end

with a negative result, there the plain, sound sense, and the pure

and noble personality of Socrates find again the ideals of morality

and science.

The strong impression which the teaching of Socrates made forced

the Sophistic activity into new lines. It followed him in the at-

tempt to gain, through scientific insight, sure principles for the

ethical conduct of life. While the old schools had for the most part

become disintegrated, and had diverted their activity to the teaching

of rhetoric, men who had enjoyed intercourse with the Athenian

sage now founded new schools, in whose scientific work Socratic

and Sophistic principles were often strangely intermingled, while

the exclusively anthropological direction of their investigation

remained the same.

Among these schools, called for the most part " Socratic," though

not quite accurately, the Megarian, founded by Euclid, fell most

deeply into the unfruitful subtleties of the later Sophists. Con-

nected with this is the Elean-Eretrian School, the most unimportant.

The fundamental contrast, however, in the conception of life which

prevailed in the Greek life of that day, found its scientific expression

in the teachings of those two schools whose opposition permeates all

ancient literature from that time on: namely, the Cynic and the

Cyrenaic, the precursors of the Stoic and Epicurean. The first of

these schools numbers among its adherents, besides its founder

Antisthenes, the popular figure of Diogenes. In the latter, which is

also called the Hedonistic School, the founder, Aristippus, was suc-

ceeded by a grandson of the same name, and later by Theodorus,

Anniceris, Hegesias, and Euemerus.

The wandering teachers known as the Sophists came in part from the earlier

scholastic societies. In the second half of the fifth century these had for themost part disappeared, and had given place to a freer announcement of opinionsattained, which was not unfavourable to special research, particularly physiologi-

cal research, as in the case of Hippo, Cleidemus, and Diogenes of Apollonia,

but which was attended by a crippling of general speculation. Only the schoolof Abdera and the Pythagorean School survived this time of dissolution. Asociety of Heracliteans which maintained itself in Ephesus appears soon to havefallen away into the pursuits of the Sophists, as in the case of Cratylus. 1

From the Atomistic School came Protagoras of Abdera (about 480-410). Hewas one of the first, and rightly the most renowned, of these wandering teachers.

Active at various times in Athens, he is said to have been convicted of impietyin that city, to have fled because of this, and to have met his death in flight. Ofhis numerous treatises, grammatical, logical, ethical, political, and religious in

their character, very little has been preserved.

1 In Plato (Thecet. 181 A) they are called ol ptovres: cf. Arist. Met. IV. 5.

1010 a 13.

Page 91: Windelband, History of Philosophy, Vol. 1

Chap. 2.] The Anthropological Period. 71

Gorgias of Leontini (483-375) was in Athens in 427 as an envoy from hia

native city, and there gained great literary influence. In old age he lived in

Larissa in Th.essa.ly. He came from the Sicilian school of orators, with whichEmpedocles also had been connected. 1

Concerning Hippiaa of Elis, with the exception of some opinions (amongwhich are those criticised in the Platonic dialogue Hippias Major"), it is knownonly that he made great parade of his "much knowledge." Of Prodicus of

Iulis, a town on the island of Ceos, the familiar allegory " Hercules at the Cross-

roads" is preserved by Xenophon, Memor. II. 1,21. The remaining Sophists,

known for the most part through Plato, are without intrinsic importance. Weknow only that this or that characteristic affirmation is put in the mouth of oneor another.

In forming a conception of the Sophistic doctrine we have to contend with the

difficulty that we are made acquainted with them almost exclusively throughtheir victorious opponents, Plato and -Aristotle. The first has given in the Pro-tagoras a graceful, lively delineation of a Sophist congress, redolent with fine

irony, in the Gorgias a more earnest, in the Thewtetus a sharper criticism, andin the Cratylus and Euthydemus supercilious satire of the Sophists' methods of

teaching. In the dialogue the Sophist, to which Plato's name is attached, anextremely malicious definition of the theories of the Sophists is attempted, andAristotle reaches the same result in the book on the fallacies of the Sophists

(Ch. I. 166 a 21).The history of philosophy for a long time repeated the depreciatory judg-

ment of opponents of the Sophists, and allowed the word cro<pi<TTT]s (whichmeant only a "learned man," or, if you will, a "professor") to bear the dis-

paraging meaning which they had given it. Hegel rehabilitated the Sophists,

and thereupon it followed, as often happens, that they were for a time over-

estimated, as by Grote.M. Schanz, Die Sophisten (Göttingen, 1867).

Socrates of Athens (469-399) makes an epoch in the history of philosophy,

even by his external characteristics, by his original personality, and his newstyle of philosophising. He was neither savant nor wandt ring teacher, be-

longed to no school and adhered to none. He was a simple man of the people,

the son of a sculptor, and at first busied himself with the chisel. In his ardentdesire for knowledge he absorbed the new doctrines with which the streets of

his native city re-echoed, but did not allow himself to be dazzled by these brill-

iant rhetorical efforts, nor did he find himself much advanced by them. Hiskeen thought took note of their contradictions, and his moral earnestness wasoffended by the superficiality and frivolity of this constant effort after culture.

He held it to be his duty to enlighten himself and his fellow-citizens concerningthe emptiness of this pretended knowledge, and, through earnest investigation,

to follow after truth. So, a philosopher of this opportunity and of daily life, heworked unremittingly among his fellow-citizens, until misunderstanding and per-

sonal intrigue brought him before the court which condemned him to the deaththat was to become his greatest glory.

The accounts concerning him give a clear and trustworthy picture of his per-

sonality. In these accounts Plato's finer and Xenophon's coarser portrayal

supplement each other most happily. The first in almost all his writings brings

out the honoured teacher with dramatic vividness. Of the second we have to

consider the Memorabilia (' Avo/j-uri^ovev/jiaTa ?.wKpäTovs) and the Symposium.As regards his teaching, the case is more difficult, for here the presentations of

both Xenophon and Plato are partisan writings, each laying claim to the famous^ame for his own doctrine (in the case of Xenophon a mild Cynicism). Thestatements of Aristotle are authoritative on all essential points, because of thegreater historical separation and the freer point of view.

E. Albert!, Sokrates (Göttingen, 1869) ; A. Labriola, La Dottrina di Socrate(Naples, 1871) ; A. Fouillee, La Philosophie de Socrate (Paris, 1873).

Euclid of Megara founded his school soon after the death of Socrates. Thetwo Eristics (see below), Eubulides of Miletus, Alexinus of Elis, DiodorusCronua of Caria (died 307), and Stilpo (380-300), are to be mentioned as

1 In regard to these relationships cf. H. Diels, Berichte der Berl. Akademie.1884, pp. 343 ff.

Page 92: Windelband, History of Philosophy, Vol. 1

72 The Greeks : Anthropological Period. [Part L

belonging to this school, which had only a brief existence, and later becameincorporated with the Cynics and Stoics. The same is true of the society whichPheedo, the favourite pupil of Socrates, founded in his home at Elis, and whichMenedemus soon after transplanted to Eretria. Cf. E. Mallet, Histoire deVecole de Megäre et des ecoles d'Elis et d"

1 Eretrie (Paris, 1845).

The founder of the Cynic School (named after the gymnasium Cynosar-ges) was Antisthenes of Athens, who, like Euclid, was an older friend of

Socrates. The singular Diogenes of Sinope is rather a characteristic by-figure

in the history of civilisation than a man of science. In this connection Cratesof Thebes may also be mentioned. Later this school was blended with that of

'he Stoics.

F. Diimmler, Antisthenica (Halle, 1882) ; K. W. Göttling, Diogenes der

KyniJcer, oder die Philosophie des griechischen Proletariats (Ges. Abhandl.I. 251 ff.)-

Aristippus of Cyrene, a Sophist and wandering teacher, somewhat youngerthan Euclid and Antisthenes, and united only for a little time with the Socratic

circle, founded his school in old age, and seems to have left to his grandson thesystematic development of thoughts, which, for himself, were rather a practical

principle of life. The above-named successors (Theodoras, etc.) extend into

the third century, and form the transition to the Epicurean School, which tookup the remnants of the Hedonistic into itself.

A. Wendt, De Philosophia Cyrenaica (Göttingen, 1841).

§ 7. The Problem of Morality.

The reflections of the Gnomic poets and the sentences of the

so-called seven wise men had already, as their central point, the

admonition to observe moderation. In like manner the pessimistic

complaints which we meet among poets, philosophers, and moralists

of the fifth century are directed for the most part against the

unbridled license of men, their lack of discipline and of obedience

to law. The more serious minds discerned the danger which the

passionate seething and foaming of public life brought with it, and

the political experience that party strife was ethically endurable

only where it left the order of the laws untouched, made subjection

to law appear as the supreme duty. Heraclitus and the Pythagoreans

expressed this with complete clearness, and knew how to attach it

to the fundamental conceptions of their metaphysical theories. 1

We meet here with two assumptions which even among these

thinkers appear as self-evident presuppositions. The first is the

validity of laws. The nai've consciousness obeys the commandwithout asking whence it com^s or by what it is justified. Laws

have actual existence, those of morals as well as those of the courts

;

they are here once for all, and the individual has to follow them.

No one in the pre-Sophistic period thought of examining the law

and asking in what its claim to valid authority consists. The sec-

ond assumption is a conviction which is fundamental in the moralis-

ing of all peoples and all times : viz. that obedience to the law

brings advantage, disregard of it, disadvantage. As the result of

1 Cf. above, p. 63, note 5.

Page 93: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 7.] Problem of Morality : the Sophists. 73

this thought admonition takes on the character of persuasive coun-

sel,1 which is directed to the shrewdness of the one admonished as

well as to the desires slumbering within him.

With the Greek Enlightenment confidence in both of these pre-

suppositions began to waver, and accordingly morality became for

it a problem.

1. The impulse to this came from the experiences of public life.

The frequent and sudden change of constitutions was indeed adapted

to undermine the authority of law. It not only took away the halo

of unconditional, unquestioned validity from the individual law,

but it accustomed the citizen of the democratic republic especially

to reflect and decide upon the ground and validity of laws as he

consulted and voted. Political law became a subject for discussion,

and the individual set himself with his judgment above it. If, now,

besides noting this mutation in time, attention is also given

to the variety exhibited not only in the political laws, but also in

the usages prescribed by customary morality in the different states

and among different peoples, the consequence is that the worth of

universal validity for all men can no longer be attributed to laws.

At least this holds good in the first place for all laws made by man;in any case, therefore, for political laws.

In the face of these experiences the question arose whether there

is anything whatever that is valid everywhere and always, any law

that is independent of the difference between peoples, states, and

times, and therefore authoritative for all. Greek ethics began thus

with a problem which teas comjrtetely parallel to the initial problem of

physics. The essence of things which remains ever the same and

survives all changes the philosophers of the first period had called

Nature (<£ixris) :

3it is now asked whether there is also determined

by this unchanging Nature (<£ixm) a law that is exalted above

all change and all differences, and in contrast with this it is pointed

out that all existing prescriptions valid only for a time, and within

a limited territory, are given and established by human institution or

statute (Secret or vo/aw).

This contrast between Nature and institution or statute is the

most characteristic work of the Greek Enlightenment in the forma-

1 A typical example of this is the allegory of Prodicus, in which the choosingHercules is promised golden mountains by Virtue as well as by Vice, in case hewill intrust himself to her guidance.

2 Ilippias in Xen. Mem. IV. 4, 14 ff.

3 Uepl Screws is the title borne by the writings of all the older philosophers.

It is to be emphasised that the constitutive mark of the concept <f>v<rt% wasoriginally that of remaining ever like itself. The contrary of this is then the

transient, that which occurs a single time.

Page 94: Windelband, History of Philosophy, Vol. 1

74 The Greeks : Anthropological Period. [Part I.

tion of conceptions. It dominates the entire philosophy of the

period, and has from the beginning not only the meaning of a prin-

ciple of genetic explanation, but the significance of a norm or stan-

dard for the estimation of worth. If there is anything universally

valid, it is that which is valid " by Nature " for all men without

distinction of people and time ; what has been established by manin the course of history has only historical worth, worth for a single

occasion. That only is justly authorised which Nature determines,

but human institution goes beyond this. The " law " (vo/xos) tyr-

annises over man and forces him to much that is contrary to Nature. 1

Philosophy formulated in its conceptions that opposition between a

natural, " divine " law and the written law, which formed the theme

of the Antigone of Sophocles.

Out of this antithesis came the problems, on the one hand, to

establish in what this law of Nature, everywhere the same, consists;

on the other, to understand how, in addition to this, the institutions

of historical law arise.

The first problem Protagoras did not avoid. In the mythical

presentation of his thought which Plato has preserved,2 he taught

that the gods gave to all men in equal measure a sense ofjustice, and

of ethical respect or reverence (SUrj and aiSws), in order that in the

struggle of life they might be able to form permanent unions for

mutual preservation. He found, therefore, the <j>v<n<s of practical

life in primary ethical feelings which impel man to union in society

and in the state. The carrying out of this thought in its details and

the definition of the boundary between this which is valid by Nature

(<J3va€i) and the positive determinations of historical institution are

unfortunately not preserved to us.

There are, however, many indications that the theory of the

Sophists proceeded from such fundamental conceptions to a wide-

reaching criticism of existing conditions, and to the demand for pro-

found revolutions in social and political life. The thought was

already at that time forcing its way forward, that all distinctions

between men before the law rest only upon institution, and that

Nature demands equal light for all. Lycophron desired to do away

with the nobility. Alcidamas 3 and others 4 combated slavery from

this point of view. Phaleas demanded equality of property as well

as of education for all citizens, and Hippodamus was the first to

1 Hippias in Plat. Prot. 337 C.2 Plat. Prot. 320 ff. Cf . A. Harpff, Die Ethik des Protagoras (Heidelberg,

1884).3 Arist. Phet. I. 13, 1373 b 18. Cf. also Orat. Attic, (ed. Bekker) II. 154.

* Arist. Pol. I. 3, 1253 b 20.

Page 95: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 7.] Problem of Morality : the Sophists. 75

project the outlines of an ideal state, constituted according to

reason. Even the thought of a political equality of women with

men came to the surface in this connection. 1

If now positive legislation deviates from these demands of Nature,

its rationale is to be sought only in the interests of those who makethe laws. Whether this takes the form assumed in the opinion of

Thrasymachus 2 of Chalcedon, who held that it is those in powerwho by means of the law force the subjects to do what is for their

(the masters') advantage, or whether it wears the contrary form as

developed by Callicles,3 that laws have been erected by the great

mass of the weak as a bulwark against the power of strong person-

alities which would be superior to the individual, and that according

to the view of Lycophron 4all those who do no harm to others thus

mutually assure for themselves life and property,— in all these

cases the ground of the laws lies in the interests of those who makethem.

2. If personal interest is therefore the ground for setting up laws,

it is also the sole motive for obeying them. Even the moralist wishes

to convince man that it is for his interest to accommodate himself

to the law. From this it follows, however, that obedience to the

law is under obligation to extend only so far as it is the indi-

vidual's interest. And there are cases where the two do not coincide.

It is not true that only subordination to law makes a man happy

;

there are great criminals, so Polus works out the thought,5 whohave attained the happiest results by the most frightful misdeeds.

Experience contradicts the claim that only right doing leads to

happiness ; it shows rather that a shrewd conduct of life, restrained

by no regard for right and law, is the best guaranty of good for-

tune. 6

Through such considerations the scepticism which had originally,

as it seems,7 been directed only toward the validity of political

law, gradually attacked that of the moral laws as well. WhatPolus, Callicles, and Thrasymachus propound in the Platonic dia-

logues, the Gorgias and the Republic, with regard to the concep-

tions of the just and unjust (6Ykcuoi/ and olSlkov) has reference in

equal measure to the moral and to the political law. This double

reference is effected through the middle ground of the characteristics

1 The persiflage in the Ecclesiazusce of Aristophanes can refer only to this.2 Plat. Bpp. 338 C.8 Plat. Gory. 483 B.* Arist. Pol. III. 9, 1280 b 11.5 In Plat. Gorg. 471.6 Cf. the praise of &8iida by Thrasymachus in Plat. Rep. 344 A.T

This is especially true of Protagoras, perhaps also of Ilippias.

Page 96: Windelband, History of Philosophy, Vol. 1

76 The Greeks : Anthropological Period. [Part L

of penal justice, and proves that the law of Nature is set over

against, not only the civil law, but also the requirements of morals.

In both respects the naturalism arid radicalism of the younger

Sophists pushed on to the extreme consequences. The weak maysubject himself to the law ; he is, though, but the stupid man, serv-

ing the uses of others by so doing ;

' the strong, however, who is at

the same time the wise, does not allow himself to be led astray by

the law ; he follows solely the impulse of his oivn nature. And this

is the right, if not according to human law, yet according to the

higher law of Nature. She shows in all living beings that the

stronger should rule the weaker ; only for the slave is it becoming

to recognise a command above himself. The free man should not

bridle his desires, but let them have full development ; according to

human law it may be a disgrace to do injustice, according to the

dictates of Nature it is a disgrace to suffer injustice. 2

In such forms the individual's natural disposition, the constitution

of his impulses, was proclaimed as law of Nature, and exalted to be

the supreme laiv of action ; and Archelaus, a disciple of Anaxagoras,

belonging to the Sophistic period, proclaimed that the predicates

good and bad, "just" and "shameful" (Si'kcuov— ato-^pov), spring

not from Nature, but from Institution. All ethical judging is con-

ventional}

3. Religious ideas were also involved in this overthrow as a mat-

ter of course, and all the more since after their theoretical value

had been taken away, at least in educated circles, by the cosmologi-

cal philosophy typified by Xenophanes, they had retained recogni-

tion only as allegorical methods of presenting ethical conceptions.

In this latter line of thought the school of Anaxagoras had been

active for a time, especially a certain Metrodorus of Lampsacus. It

was only a consequence of the ethical relativism of the Sophists

when Prodicus taught that men had made to themselves gods out of

all that brought them blessing, and when Critias declared belief in

the gods to be an invention of shrewd statecraft. 4 If such claims

still excited indignation among the masses and the powers of the

official priesthood,5it was easy for Protagoras in the presence of

these questions to wrap himself in the mantle of his scepticism. 6

4. The position of Socrates with reference to this whole move-

ment presents two sides : on the one hand, he brought the principle

1 Thrasymachus in Plat. Rep. 343 C.2 Callicles in Plat. Gorg. 483 A and 491 E.3 Diog. Laert. II. Ki* Sext. Emp. Adv. Math. IX. 51-54.5 As is shown by the condemnation of Diagoras of Melos (Aristoph. Av. 1073).6 Diog. Laert. IX. 51.

Page 97: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 7.] Problem of Morality : Socrates. 77

underlying the movement to its clearest and most comprehensive

expression ; on the other hand, he set himself in the most vigorous

manner against its outcome, and both these sides of his activity,

contrary as they seem to be and much as this external opposition

had to do with the tragic fate of the man, stand, nevertheless, in the

most exact and rigidly consistent connection ; for just by grasping

the principle of the Enlightenment in all its depth, and formulating

it in its full force, did Socrates succeed in developing from it a

positive result of wide-reaching power.

For him, also, the time for following traditional customs without

question is past. Independent judgment of individuals has taken

the place of authority. But while the Sophists gave their attention

to the analysis of the feelings and impulses which lie at the basis of

the actual decisions of individuals, and ultimately saw themselves

forced to adjudge to all these motives the equal right of an unfold-

ing in accordance with the necessity of Nature, Socrates, on the

contrary, reflected upon precisely that element which was the deci-

sive factor in the culture of his time : namely, the practical, polit-

ical, and social significance which knowledge and science had

achieved. Just through the process in which individuals had

achieved independence, through the unfettering of personal passions,

it had become evident that in all fields man's ability rests upon his

insight. In this Socrates found that objective standard for the esti-

mation of men and their actions which the Sophists had sought in

vain in the machinery of feelings and desires.

Ability, then, or excellence (Tüchtigkeit, aperrj) is insight. Hewho acts according to feelings, according to presuppositions that

are not clear, according to customs that have been handed down,

may indeed occasionally hit the right thing, but he does not knowit, he is not sure of the issue ; he who is entirely involved in delusion

and error as to the matter in hand is certain to make mistakes ; he

only will be able to act right who has the right knowledge of things

and of himself. 1 Scientific knowledge (eVtcrT^ja^) is therefore the

basis of all qualities which make man able and useful, of all single

dpeTai.

This insight consists, on the one hand, in an exact Jcnoirfedge of the

things to which the action is to relate. Man should understand his

business ; as we find the able man in every business to be the one

who has learned it thoroughly and knows the objects with which he

has to work, so should it be also in civil and political life ; here, too,

1 These fundamental thoughts of Socrates are reproduced by Xenophon andP>ato in countless turns and variations. In Xenophon the passage, Mem. III.

ch. 9, is most important for comparison; in Plato, the dialogue ^ntafforas.

Page 98: Windelband, History of Philosophy, Vol. 1

78 The Grreeks : Anthropological Period. [Part 1.

only insight should be trusted. 1 The individual excellences differ-

entiate themselves accordingly with reference to the objects which

the knowledge concerns in the individual case

;

2 common to all,

however, is not only knowledge in general, but also self-knowledge.

Hence Socrates declared it to be his principal vocation to educate

himself and his fellow-citizens to earnest self-examination; the

yvwOi aeavTov was the watch-word of his teaching.3

5. These considerations, which Socrates developed out of the

principles by which practical ability or excellence is determined,

became transferred by the aid of the ambiguity in the icord apcrr},4

to ethical excellence also, or virtue, and so led to the fundamental

doctrine that virtue consists in knowledge of the good.* So far the

course of thought followed by Socrates is clear and free from doubt.

The sources become less clear when we ask what the man who was

so strenuous to reach clearly defined conceptions intended by the

good. According to Xenophon's exposition, the good (dyaOov) must

have coincided everywhere, for his master, with the profitable or

useful (axfreXi/xov) . Virtue would then be the knowledge of what

was suited to the end in view, or useful, in each particular instance.

This interpretation is the easiest to attach to that analogy between

moral virtue and the various kinds of excellence shown in daily

life, which Socrates really taught, and the presentation given in the

earliest Platonic dialogues, in particular the Protagoras attributes

to Socrates this standpoint of individual advantage. Insight or dis-

cernment (here called prudence, <f>p6vrj<ri<s) is a measuring art, which

weighs exactly the benefit and the harm that will result from the

action, and so chooses what is most to the purpose. In further agree-

ment with this view is the fact that in exact contrast with the

Sophists, who demanded a free and uncramped development of the

passions, Socrates emphasised no virtue so much, and exhibited none

so fully in his own life, as that of self-control (<jo)<f>oo-uvr)) .

But according to this interpretation the Socratic conception of

the good would be indefinite in its content ; decision must be madefrom case to case as to what suits the end in view, or is useful, and

1 Hence, too, the anti-democratic position, so fatal for his personal destiny,taken by Socrates, who demanded expressly that the most difficult and mostresponsible art, that of governing, should be practised only by those of the mostcomplete discernment, and who on this account absolutely rejected the appoint-ment of state officials by lot or popular choice.

2 Socrates did not attempt a system of the individual excellences ; on theother hand, he did give by way of example definitions of courage (cf. the PlatonicLaches), piety (Plat. Euthyphro, Xen. Mem. IV. 6, 3), justice {Mem IV. 6, 6),etc.

3 As defined by his theoretical philosophy; see § 8.4 The same ambiguity which has given occasion to countless difficulties lies

in the Latin virtus ; so, too, in dyadbv, bonum, good.

Page 99: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 7.] Problem of Morality : Socrates. ?y

instead of the good we should again always have what is good for

something. 1 It may be regarded as certain that Socrates strove to

transcend this relativism, and also that by reason of the anthropo-

logical basis of his thinking he did not get beyond this position in

the formulation of his conceptions. His doctrine that it is better to

suffer wrong than to do wrong, his strict conformity to law, in

accordance with which he scorned to avoid the execution of an

unjust sentence and preserve himself by flight for further life and

activity, his admonition that the true meaning of life consists in

tvTvpa&a, in continual right-doing, in man's ceaseless labour for ethical

improvement, in the participation in all that is good and beautiful

(KaXoxayaOia), especially, however, his erotic, i.e. his doctrine that

friendship and the relation of attachment between teacher and

taught should consist only in a mutual striving to become good or

constantly better through their life in common and their mutual

furtherance of each other's aims,— all this goes far beyond the con-

ception presented by Xenophon. It can be united with the stand-

point of utility only if we attribute to Socrates the distinction

between the true welfare of the soul, on the one hand, and earthly

gain, on the other, which Plato makes him set forth in the Phceclo,

but of which we elsewhere find but slight traces, since the historic

Socrates, even according to Plato's Apology, maintained a completely

sceptical position with regard to personal immortality, and did not

know the sharp Platonic separation between immateriality and cor-

poreality. Socrates teaches, indeed, even according to Xenophon,

that man's true fortune is to be sought, not in outward goods nor in

luxurious life, but in virtue alone : if, however, this virtue is to

consist only in the capacity to recognise the truly useful and act

accordingly, the doctrine moves in a circle as soon as it maintains

that this truly useful is just virtue itself. In this circle Socrates

remained fast; the objective determination of the conception of the

good which he sought he did not find.

6. However indefinite the answer to the question as to whatshould properly form the content of that knowledge of the goodwhich constitutes virtue, Socrates was at all events convinced—and this proved much more important— that this knoicledge is

in itself sufficient to cause one to do the good, and so bring happi-

ness. This proposition, which may serve as a type of a rationalis-

tic conception of life, contains two pregnant presuppositions, one

psychological, viz. pronounced intellectualism, the other ethical, viz.

pronounced eudozmonism.

i Xen. Mem. III. 8, 5.

Page 100: Windelband, History of Philosophy, Vol. 1

80 The Greeks : Anthropological Period. £Part i-

The fundamental assumption which Socrates thus makes is

indeed the expression of his own reflective, judicious nature. Every

man, he says, acts in the manner that he considers best suited for

his end, most beneficial and most useful ; no one does that which

he knows to be unfit for the end in view, or even fit in a lesser

degree. If, then, virtue is knowledge of what is to the purpose, it

follows immediately that the virtuous man acts in accordance with

his knowledge, therefore to the purpose, rightly, in the way that is

beneficial to him. No one does wrong knowingly and purposely : he

only does not act rightly who has not right insight. If it sometimes

seems as if some one acted wrongly in the face of better insight —" against his better judgment "— it must be that he was not clearly

and surely in possession of this better knowledge, for otherwise he

would have purposely injured himself, which is absurd.

In this a fundamental difference between Socrates and the

Sophists becomes evident: the latter maintained the originality of

the will, and on that account its warrant from Nature ; for Socrates,

to will a thing and to regard a thing as good, profitable, and useful

are the same thing. Knowledge determines the will without

opposition ; man does what he holds to be best. True as it may be

that Socrates was in error in this opinion, and that the truth lies in

the mean between him and the Sophists, this his intellectualistic

conception of the will came to exercise a decisive influence over all

ancient ethics.

Sin is, then, error. He who does a bad act does it from a mistaken

judgment, regarding the bad, i.e. the injurious, as the good ; for every

one believes that he is doing the good, i.e. the advantageous. Only

because the case stands thus is there any meaning in instructing

men ethically ; only for this reason is virtue capable of being taught.

For all teaching addresses itself to man's knowledge. Because mancan be taught what the good is, therefore— and by this means alone

1— he can be brought to the stage of right action. Were virtue not

knowledge, it would not be capable of being taught.

From this standpoint Socrates raised the customary morality

taught by the popular moralising to a scientific plane. All his

keenness, his subtlety, and dialectical dexterity were employed l to

prove against the Sophists that not only the surest, but even the

only sure way of attaining to permanent happiness, lies in obeying

ethical prescriptions under all circumstances, in subordination to law

and morals. So he gives back to Authority her right. The prin-

1 Compare in Plato the refutation of Thrasymachus in the first book of the

Republic, which may be regarded as Socratic in its principles, but which in part

« very weakly supported, both in form and in matter.

Page 101: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 7.] Problem of Morality : Socrates. 81

ciple of the Enlightenment tolerates no unquestioning subjection to

the existing state of things and requires examination of the laws;

but these laws sustain the examination, they evince themselves to be

requirements made by insight into what is for the best ; and because

it has now been recognised that it is the right course to obey them,

unconditional obedience must be rendered. 1 Far from being in con-

flict with the institutions of law and morals, Socrates is rather the

one who undertook to prove their reasonableness and thereby their

claim to universal validity.2

F. Wildauer, Socrates'1 Lehre vom Willen. Innsbruck, 1877.

M. Heinze, Der Eudämonismus in der griechischen Philosophic Leips.

1883.

7. In addition to the psychologico-ethical presuppositions that

the will is always directed toward what is recognised as good,

and that therefore virtue, as knowledge of the good, draws after it

of itself the appropriate action, we find in the argumentations of

Socrates the further opinion that this appropriate action of the

virtuous man actually attains its end and makes him happy. Hajypi-

ness or well-being (euSai/xovia) is the necessary result of virtue. Theintelligent man knows, and hence does, what is good for him ; he

must then, through his doing, become happy also. This assump-

tion applies, however, only to a perfect intelligence which would

be absolutely certain of the effects that an intended action would

have in the connected series of the world's events.

1 In details, as might be expected from the nature of the case, this rehabilita-

tion of the popular morals falls into trivial moralising, especially as Xenophonportrays it. But while Socrates hoped precisely by this means to render the

right service to his people, it proved to be just the point where he came to the

ground between two stools : with the Sophists and their adherents, he passed for

a reactionary ; on the other hand, the men who, like Aristophanes, saw pre-

cisely in the questioning of the authority of law and morals in general, the dan-gerous cancer of the time, without investigation classed him who wished to

place this authority on a basis of reason, among those who were underminingit. So it was that it could come about that Socrates appeared in the Clouds of

Aristophanes as the type of Sophistic teaching which he combated.2 It is hence quite alien to the principles of Socrates to demand or even to

allow for every individual act a special examination of the grounds of the polit-

ical or ethical command. If, for example, it has once been recognised as right

to obey the ordinances of the government under all circumstances, this obedience

must then be rendered, even if the ordinance evidently commands the unreason-

able and the unjust ; cf . Plato's Crito. If, as was true of Socrates himself, a manis convinced that his life is under divine guidance, and that where his insight

does not suffice, a higher voice warns him through his feeling, — at least, warnshim away from what is wrong, — then he must obey this voice. Cf. on the

Salfxoviov,' § 8. The essential thing always is that a man give an account to him-self of his doing, but the grounds on which he acts in so doing may even consist

in such maxims as exclude an examination in individual cases.

Page 102: Windelband, History of Philosophy, Vol. 1

82 The Greeks : Anthropological Period. [Part 1.

The transmitted expressions of Socrates, in fact, make the impres-

sion that he was convinced that man could possess that insight

which by its operation upon his action and its consequences is

adapted to bring about happiness, and that he might gain this

insight through philosophy : that is, through unremitting earnest

examination of himself, of others, and of the relations of humanlife. Investigations as to how far the world's course, which mancannot foresee, may cross and destroy the operation even of the best

planned and most intelligent conduct of life, are not to be pointed

out in the teaching of Socrates. When we consider the slight

degree of confidenc, vvhich he otherwise had in human knowledge,

as soon as this attempted to venture beyond establishing ethical

conceptions and practical requirements, we can explain the above

conviction only on the following basis— he did not fear that the

providential guidance, which was for him indeed an object not of

knowledge, but of faith, would frustrate the beneficial consequences

of right action.

8. Socrates had defined virtue, the fundamental ethical concep-

tion, as insight, and this in turn as knowledge of the good, but had

given to the concept of the good no universal content, and in a cer-

tain respect had left it open. This made it possible for the most

diverse conceptions of life to introduce their views of the ultimate

end (re'Xos) of human existence into this open place in the Socratic

concept ; and so this first incomplete work in the formation of ethi-

cal conceptions at once afforded the material for a number of partic-

ular structures. 1 The most important of these are the Cynic and

the Cyrenaic. Both present the attempt to define the true intrinsic

worth of the life of the individual in a universal manner. Both

wish to show in what man's true happiness consists, how man must

be constituted and how he must act in order to attain this with cer-

tainty ; both call this constitution or disposition through which

participation in happiness is gained, virtue. The eudamonistic side

of the Socratic ethics is here developed in an entirely one-sided

manner, and though universal validity is vindicated for the concep-

tion proposed, the point of view of the individuaVs happiness forms so

exclusively the standard that the worth of all relations of public

life even is estimated by it. In Cynicism, as in Hedonism, the Greek

spirit is proceeding to appropriate the fruit which the conditions

1 So indeed in the case of Xenophon and iEschines ; the philosophising cob-bler Simon, too, seems to have have been thus dependent on Socrates. Whatthe Megarian and the Elean-Eretrian schools accomplished in this respect is

too indefinitely transmitted to us, and is too closely in contact with Cynicism,to deserve separate mention.

Page 103: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 7.] Problem of Morality : Antisthenes. 83

of life brought about by civilisation yield for the fortune of the

individual. The criticism of the social conditions and authorities,

begun by the Sophists, has won a fixed standard through the medi-

ating aid of the Socratic conception of virtue.

The doctrine of virtue taught by Antisthenes 1 takes at the begin-

ning a high and specious turn at the point where the doctrine finds

itself hopelessly entangled in the Socratic circle. He declines to

define more closely the contents of the concept of the good, anddeclares virtue itself to be not only the highest, but the only good,

understanding, however, by virtue essentially only the intelligent con-

duct of life. This alone makes happy, not indeed through the conse-

quences which it brings about, but through itself The contentment

that dwells within the right life itself is accordingly completely

independent of the world's course : virtue is itself sufficient for

happiness; the wise man stands free in the presence of fate and

fortune.

But this Cynic conception of virtue as sufficient in itself is, as is

shown by its further development, in nowise to be interpreted as

meaning that the virtuous man should find his fortune in doing

good for its own sake amid all the whims of fate. Cynicism did

not rise to this height, however much it may sound like it whenvirtue is celebrated as the only sure possession in the vicissitudes

of life, when it is designated as the only thing to be striven for,

and baseness, on the contrary, as the only thing to be avoided. This

doctrine is a postulate derived with great logical consistency fromthe Socratic principle that virtue necessarily makes happy (cf.

above, 7), and from this postulate Antisthenes sought in turn to

define the real contents of the concept of virtue.

If, namely, virtue is to make happy with certainty and under all

circumstances, it must be that conduct of life which makes man as

independent as possible of the course of events. Now every want andevery desire is a bond which makes man dependent upon fortune,

in so far as his happiness or unhappiness is made to consist in

whether a given wish is fulfilled or not by the course of life. Wehave no power over the outer world, but we have power over our

desires. We expose ourselves the more to alien powers, the morewe desire, hope, or fear from them ; every desire makes us slaves of

the outer world. Virtue, then, which makes man independent, canconsist only in suppression of desires, and restriction of wantsto the smallest conceivable measure. Virtue is freedom fromwants. 2— from the standpoint of eudaernonism certainly the most

1 Principally preserved in Diog. Laert. VI. 2 Xen. Symp. 4, 34 fL

Page 104: Windelband, History of Philosophy, Vol. 1

84: The Greeks: Anthropological Period. [Part L

consistent conclusion, and one that must have appealed especially to

men of a humble position in life such as we find the Cynics to be in

part.

By carrying out this thought in a radical manner the Cynics came

to occupy a purely negative attitude toward civilisation. By aiming

to reduce the measure of the virtuous wise man's wants to what was

absolutely inevitable, and to regard all other strivings as pernicious

or indifferent, they rejected all the goods of civilisation and attained

the ideal of a state of Nature, — an ideal stripped of all higher worth.

Taking up earlier Sophistis theories and developing them farther,

they taught that the wise man accommodates himself only to what

Nature peremptorily demands, but despises all that appears desir-

able or worthy of obedience merely as the result of human opinion

or institution. Wealth and refinement, fame and honour, seemed to

them just as superfluous as those enjoyments of the senses which

went beyond the satisfaction of the most elementary wants of hunger

and love. Art and science, family and native land, were to them

indifferent, and Diogenes owed his paradoxical popularity to the

ostentatious jest of attempting to live in civilised Greece as if in a

state of Nature, solely tftwret.

In this way the philosophising proletarian forced himself to despise

ail the good things of civilisation, from the enjoyment of which he

found himself more or less excluded. On the other hand, he recog-

nised none of the laws to which civilised society subjected itself, as

binding in themselves, and if there is any truth at all in the coarse

anecdotes which antiquity relates on the subject, this class took

pleasure in scoffing openly at the most elementary demands of

morals and decency. This forced and, in part, openly affected nat-

uralism knows nothing any longer of Si/07 and aiSws (justice and rev-

erence), which the older Sophistic* teaching had allowed to remain

as natural impulses, and elicits a conception of virtue which sup-

poses that greed and lust complete the essential qualities of the

natural man.

Yet the Cynics were not so bad as they made themselves.

Diogenes even preserved a remnant of respe<?t for mental training,

as the only thing which could free man from the prejudices of con-

ventional institutions and lead to freedom from wants by insight

into the nothingness of the pretended goods of civilisation. Healso conducted the education of the sons of Xeniades, a Corinthian

Sophist, according to the principles of the Cynic naturalism, and

not without success.

On the whole, this philosophy is a characteristic sign of the time,

the mark of a disposition which if not hostile, was yet indifferent

Page 105: Windelband, History of Philosophy, Vol. 1

Cha^-. 2, § 7.j Problem of Morality : Aristippus. 85

to society and had lost all comprehension of its ideal goods ; it ena-

bles ns to see from within how at that time Greek society was dis-

integrating into individuals. When Diogenes called himself a

cosmopolitan, there was in this no trace of the ideal thought of a

community of all men, but only the denial of his adherence to any

civilised community ; and if Crates taught that the plurality of gods

exists only in the opinion of men, and that, " according to Nature,"

there is but one God, there is in the Cynic doctrine no trace to war-

rant the conclusion that this monotheism was for them an especially

clear idea or even an especially deep feeling.

9. In complete contrast with this system stands Hedonism, the

philosophy of regardless enjoyment. Starting as did the Cynics

from the incompleteness of the Socratic doctrine, Aristippus struck

out in the opposite direction. He was quick to give to the concept

of the good, a clear and simple content,— that of pleasure (rjhovr)).

This latter conception at first does duty under the general psycholo-

gical meaning of the feeling of contentment which grows out of

the fulfilment of every striving and wish. 1 Happiness is then the

state of pleasure which springs from the satisfied will. If this is

the only thing to be considered, it is a matter of indifference what

the object of will and of gratification is ; all depends on the

degree of pleasure, on the strength of the feeling of satisfaction. 2

This, however, in the opinion of Aristippus, is present in the highest

degree in the case of sensuous, bodily enjoyment which relates to

the immediate present, to the satisfaction of the moment. If, then,

virtue is knowledge directed toward happiness, it must enable manto enjoy as much and as vigorously as possible. Virtue is ability

for enjoyment.

Every one, to be sure, may and can enjoy ; but only the man of

education, of intelligence, of insight— the wise man — understands

how to enjoy rightly. In this we must consider not only the

intelligent appraisal (cj>p6vr)<n<;) , which knows how to select, amongthe various enjoyments that present themselves in the course of

life, those which will afford the pleasure that is highest, purest,

least mixed with pain ; we must consider also the inner self-posses-

sion of the man who is not blindly to follow every rising appetite,

and who, when he enjoys, is never to give himself entirely up to

the enjoyment, but is to stand above it and control it. The enjo}'-

ment which makes man the slave of things is, indeed, as the Cynics

1 Besides this, also, Xenophon not infrequently puts the ijdij into the mouthof Socrates.

2 This, too, is a completely correct consequence from the eud?emonistic prin-

ciple.

Page 106: Windelband, History of Philosophy, Vol. 1

86 The Greeks : Anthropological Period. [Part I.

say, to be rejected; but to delight in pleasure and yet not give one's

self up to it is harder than to renounce it, as they do. Of this,

however, man becomes capable through right insight only. 1

On this ground the Cyrenaics, in particular the younger Aristippus

(called /x7/rpo8t'8aKTos, "mother-taught," because his grandfather's

wisdom was transmitted to him through his mother Arete), set on

foot systematic investigations as to the origin of the Trddrj, the

feelings and impulses. In a physiological psychology which was

connected with that of Protagoras (cf. below, § 8), they traced the

varieties in feeling back to states of motion in the body : to rest

corresponded indifference, to violent motion pain, to gentle motion

pleasure. Besides such explanatory theories, however, this philos-

ophy of bonvivants extended to an unprejudiced general theory

of things. For them, too, as Theodorus taught, all ethical and legal

prescriptions were ultimately merely institutions that were valid for

the mass of men; the educated man of enjoyment gives himself

no trouble about them, and enjoys things when they come into his

possession. Theodorus, who bears the surname " the Atheist," put

aside also all religious scruples which are opposed to devotion to

sensuous enjoyment, and the school also exerted itself in this

interest to strip the halo from religious faith, so far as possible, as

is proved by the well-known theory of Euemerus, who in his Upa

avaypa<j>7i undertook to trace belief in the gods back to the worship

of ancestors and veneration of heroes.

Thus the Cyrenaics ultimately agreed with the Cynics in this,

that they, too, regarded all that is fixed vofxu, i.e. by the social

convention of morals and law, as a limitation of that right to enjoy-

ment which man has by nature (cpvaei), and which the wise manexercises without troubling himself about historical institutions.

The Hedonists gladly shared the refinement of enjoyment whichcivilisation brought with it ; they found it convenient and per-

missible that the intelligent man should enjoy the honey whichothers prepared; but no feeling of duty or thankfulness boundthem to the civilisation whose fruits they enjoyed. This same con-

dition of recognising no native land, this same turning aside from

the feeling of political responsibility, which among the Cynics grew

out of despising the enjoyments of civilisation, resulted for the

Cyrenaics from the egoism of their enjoyment. Sacrifice for

others, patriotism, and devotion to a general object, Theodorus

declared to be a form of foolishness which it did not become the

wise man to share, and even Aristippus rejoiced in the freedom from

1 Cf. Diog. Laert. II. 65 ff.

Page 107: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 8.] Problem of Science : the Sophists 87

connection with any state, which his wandering life afforded him. 1

The philosophy of the parasites, who feasted at the full table of

Grecian beauty, was as far removed from the ideal meaning of that

beauty as was the philosophy of the beggars who lay at the threshold.

In the meantime, the principle of the expert weighing of enjoy-

ments contains an element which necessarily leads beyond that

doctrine of enjoyment for the moment which Aristippus preached,

and this advance was made in two directions. Aristippus himself

had already admitted that in the act of weighing, the pleasure

and pain which would in future result from the enjoyment

must be taken into account ; Theodoras found that the highest good

was to be sought rather in the cheerful frame of mind (xa-pa-) than

in the enjoyment of the moment, and Anniceris came to see that this

could be attained in a higher degree through the spiritual joys of

human intercourse, of friendship, of the family, and of civil society

than through bodily enjoyments. This knowledge that the enjoy-

ments afforded by the intellectual and spiritual aspects of civilisa-

tion are ultimately finer, richer, and more gratifying than those

of bodily existence, leads directly over into the doctrine of the

Epicureans. But, on the other hand, the Hedonistic school could

not fail ultimately to see that the painless enjoyment to which it

aimed to educate the man of culture is but a rare lot. In general,

found Hegesias, he is to be accounted as already happy who attains

the painless state, is free from actual discomfort. With the great

mass of men discomfort, the pain of unsatisfied desires, pre-

ponderates : for them it would be better, therefore, not to live.

The impressiveness with which he presented this brought him the

surname 7r£«n.0avaTos, — he persuaded to death. He is the first

representative of eudcemonistic pessimism ; with this doctrine, how-

ever, eudaemonism refutes itself. He shows that if happiness,

satisfaction of wishes, and enjoyment are to be the meaning and

end of human life, it misses this end, and is to be rejected as

worthless. Pessimism is the last but also the annihilating con-

sequence of eudaemonism,— its immanent criticism.

§ 8. The Problem of Science. 2

P. Natorp, Forschungen zur Geschichte des Erkenntnissproblems bei denAlten. Berlin, 1884.

The Sophists were teachers of political eloquence. They were

obliged in the first instance to give instruction on the nature and

i Xen. Mem. IT. 1, 8 ff.

2[ Wissenschaft. Science, as used in this section, is nearly equivalent to

"scientific knowledge." Sometimes the subjective aspect of the term is prouii'

nent, and sometimes the objective.]

Page 108: Windelband, History of Philosophy, Vol. 1

88 The Grreeks : Anthropological Period. [Part I.

right use of language. And while they were transforming rhetoric

from a traditional art to a science, they applied themselves in the

first place to linguistic researches, and became creators of grammar

and syntax. They instituted investigations as to the parts of the

sentence, the use of words, synonyms, and etymology. Prodicus.

Hippias, and Protagoras distinguished themselves in this respect;

as to the fruit of their investigations, we are only imperfectly

informed.

1. Our knowledge of their logical acquisitions, which with the

exception of a few allusions are lost, is in a still more unfortunate

condition. For, as a matter of course, the teachers of rhetoric

treated also the train of thought in discourse. This train of thought,

however, consists in proof and refutation. It was then inevitable

that the Sophists should project a theory of proof and refutation,

and there is explicit testimony to this in the case of Protagoras. 1

Unfortunately, there is no more precise information as to howfar the Sophists proceeded with this, and as to whether they

attempted to separate out the logical Forms from those elements

which belong to the content of thought. It is characteristic that

the little information which we have concerning the logic of the

Sophists relates almost without exception to their emphasising of

the principle of contradiction. To the essential nature of the advo-

cate's task, refutation was more closely related than proof. Protag-

oras left a special treatise 2 concerning Grounds of Refutation,

perhaps his most important writing, and formulated the law of the

contradictory opposite, so far, at least, as to say that there are with

reference to every object two mutually opposing propositions, and

to draw consequences from this. He thus formulated, in fact, the

procedure which Zeno had practically employed, and which also

played a great part in the disciplinary exercises of the Sophists,

indeed the greatest part.

For it was one of the main arts of these " Enlighteners " to per-

plex men as to the ideas previously regarded as valid, to involve

them in contradictions, and when the victims were thus confused,

to force them if possible, by logical consequences, real or manufac-

tured, to such absurd answers as to make them become ridiculous

to themselves and others. From the examples which Plato 3 and

Aristotle 3 have preserved, it is evident that this procedure was not

i Diog. Laert. IX. 51 ff.

2 It is probable that Kara/3d\\ovres (sc. \6yoi) and 'AvriXoytai are only twodifferent titles of this work, the first chapter of which treated truth.

3 Plato in the Evthydemus and in the Cratylus, Aristotle in the book "On the

Sophistic Fallacies."

Page 109: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 8.] Problem of Science : the Megarians. 89

always any too purely logical, but was thoroughly sophistical in the

present sense of the word. The examples show that these people

let slip no ambiguity in speech, no awkwardness in popular expres-

sion, if out of it they might weave a snare of absurdity. The

witticisms which result are often based merely upon language,

grammar, and etymology ; more rarely they are properly logical

;

quite often, however, coarse and dull. Characteristic here, too, are

the catch-questions, where either an affirmative or negative answer,

according to the customs and presuppositions of the ordinary mean,

ings of the words, gives rise to nonsensical consequences, unforeseen

by the one answering. 1

Plato has portrayed two brothers, Euthydemus and Dionysidorus,

who practised this art of logomachy or eristic, which had great

success among the Athenians who were great talkers and accus-

tomed to word-quibbling. Aside from them, it was prosecuted

principally by the Megarians, among whom the head of the school,

Euclid, busied himself with the theory of refutation. 2 His adhe-

rents, Eubulides and Alexinus, were famous for a series of such

catches, which made a great sensation and called forth a whole lit-

erature.3 Among these there are two, the " Heap " and the " Bald-

head," 4 the fundamental thought in which is to be traced back to

Zeno, and was introduced by him into the arguments by which he

wished to show that the composition of magnitudes out of small

parts is impossible. In like manner, Zeno's arguments against

motion were amplified, even if not deepened or strengthened,5 by

another Megarian, Diodorus Cronos. Unwearied in finding out such

aporim, difficulties, and contradictions, this same Diodorus invented

also the famous argument (KupieiW) which was designed to destroy

the conception of possibility: only the actual is possible; for a

possible which does not become actual evinces itself thereby to be

impossible. 6

In another manner, also, the Sophists who were affiliated with the

Eleatics, show an extreme application of the principle of contradic-

tion, and a corresponding exaggeration of the principle of identity.

Even Gorgias seems to have supported his opinion that all state-

ments are false, upon the assumption that it is incorrect to predicate

1 As a typical example, " Have you left off beating your father ?" or " Haveyou shed your horns ? "

2 Diog. Laert. II. 107.3 Cf. Prantl, Gesch. der Log, I. 33 ff.

4 Which kernel of grain by being added makes the heap ? Which hair failing

out makes the bald head ?6 Sext. Emp. Adv. Math. X. 85 ff.

6 Cic. De Fata, 7, 13.

Page 110: Windelband, History of Philosophy, Vol. 1

90 The Grreeks : Anthropological Period. /"Part I.

of any subject anything else than just this subject itself; and the

Cynics, as well as Stilpo the Megarian, made this thought their own.

There remain, accordingly, only such purely identical judgments as,

good is good, man is man, etc.1 As a logical consequence of this,

judging and talking are made as impossible as were plurality and

motion according to the Eleatic principle. As in the metaphysics

of Parmenides, the ghost of which appears occasionally both amongthe Megarians and the Cynics (cf. below, No. 5), the lack of concep-

tions of relation permitted no combination of unity with plurality

and led to a denial of plurality, so here the lack of conceptions of

logical relation made it appear impossible to assert of the subject a

variety of predicates.

2. In all these devious windings taken by the researches of the

Sophists concerning the knowing activity, the sceptical direction is

manifesting itself. If on such grounds the logical impossibility of

all formation of synthetic propositions was maintained, this showed

that knowledge itself was irreconcilable with the abstract principle

of identity, as it had been formulated in the Eleatics' doctrine of

Being. The doctrine of Parmenides had itself become ensnared

past help in the dichotomies of Zeno. This came to most open

expression in the treatise of Gorgias, 2 which declared Being, Knowl-

edge, and Communication of Knowledge to be impossible. There is

nothing; for both Being, which can be thought neither as eternal

nor as transitory, neither as one nor as manifold, and Non-being are

conceptions that are in themselves contradictory. If, however,

there were anything, it would not be knowable ; for that which is

thought is always something else than that which actually is, other-

wise they could not be distinguished. Finally, if there were knowl-

edge, it could not be taught ; for every one has only his own ideas,.

and in view of the difference between the thoughts and the signs

which must be employed in their communication, there is no guar-

anty of mutual understanding.

This nihilism, to be sure, scarcely claimed to be taken in earnest

;

even the title of the book, wept c^ucrecos r) ir€pl rov fxrj ovtos (Concern-

ing Nature, or concerning that which is not), appears like a

grotesque farce. The Rhetorician, trained to formal dexterity, whodespised all earnest science and pursued only his art of speaking,5

indulged in the jest of satirising as empty the entire labour of philos-

1 Plat. Thecet. 201 E. Cf. Soph. 251 B.2 Extracts are found partly in the third chapter of the pseudo-Aristotelian

treatise De Xenophane, Zenone, Gorgia (cf. p. 30), in part in Sext. Emp. VIL65-86.

3 Plat. Meno. 95 C.

Page 111: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 8.] Problem of Science : Protagoras. 91

ophy, and doing this ironically in the style of Zeno's pinching-mill

of contradictions. But just the facts that he did this, and that his

work found applause, show how among the men who occupied them-

selves in instructing the people, and in the circles of scientific

culture itself, faith in science was becoming lost at just the time

when the mass of the people was seeking its welfare in it. This

despair of truth is the more comprehensible, as we see how the

serious scientific investigation of Protagoras attained the same

result.

E. Laas, Idealismus und Positivismus. I. Berlin, 1880.

W. Halbfass. Die Berichte des Piaton und Aristoteles über Protagoras.Strassb. 1882.

Sattig, Der Protagoreische Sensualismus (Zeitschrift für Philosophie, vols.

86-89).

3. The germ of the doctrine of Protagoras is found in his effort

to explain the ideas of the human mind psycho-genetically . Insight

into the origin and development of ideas was absolutely necessary

for the practical aspect of a system of ethics, and particularly for

the cultivation of rhetoric. The statements, however, which the

metaphysicians had occasionally uttered, were in nowise sufficient

for the purpose, constructed as they were from general presupposi-

tions and permeated by them ; on the contrary, the observations in

physiological psychology which had been made in the more recent

circles of investigators who were more given to natural science,

offered themselves as fit for the purpose. Thinking and perceiving

had been set over against each other from the point of view of

their relative worth ; this determining element now disappeared for

Protagoras, and so there remained for him only the view of the-

psychological identity of thinking and perceiving, — a view to which

even those metaphysicians had committed themselves as soon as

they attempted to explain ideation from the world-process (cf. § 8).

In consequence of this he declared that the entire psychical life con-

sists only in perceptions} This sensualism was then illustrated by

the great mass of facts which physiological psychology had assembled

in connection with the teaching of the physicians that were scien-

tific investigators, and by the numerous theories which had been

brought forward with special reference to the process of the action

of the senses.

All these, however, had in common the idea that perception rests

in the last instance upon motion, as does every process by which

things come to be or occur in the world. In this even Anaxagoras

1 Diog. Laert. IX. 51.

Page 112: Windelband, History of Philosophy, Vol. 1

92 The Grreeks : Anthropological Period. /Tart l

and Einpedocles were at one with the Atomists, from whose school

Protagoras, as a native of Abdera, had probably gone out. This

agreement extended still farther to the assumption, made on all

sides, that in perception there was not only a condition of motion

in the thing to be perceived, but also a like condition in the percip-

ient organ. Whatever view might be taken as to the metaphysical

essence of that which was there in motion, it seemed to be acknowl-

edged as undoubted that every perception presupposed this double

motion. Empedocles had already anticipated the doctrine that the

inner organic motion advances to meet the outer. 1

On this foundation 2 the Protagorean theory of Jcnotvledge is built

up. If, that is to say, perception is the product of these two motions

directed toward one another, it is obviously something else than the

perceiving subject, but just as obviously it is something else than the

object which calls forth the perception. Conditioned by both, it is yet

different from both. This pregnant discovery is designated as the

doctrine of the subjectivity of sense-perception.

Nevertheless, in the case of Protagoras this appears with a peculiar

restriction. Since, like all earlier thinkers, he evidently could not

assume a consciousness without a corresponding existent content of

consciousness, he taught that from this double motion there was a two-

fold result : viz. perception (aladrjcns) in the man, and content ofper-

ception (to alaOnrov) in the thing. Perception is therefore indeed

the completely adequate Jcnoivledge of what is perceived, but no knowl-

edge of the thing. Every perception is then in so far true as, at

the instant when it arises, there arises also in connection with the

thing the represented content, as alaönrov, but no perception knows

the thing itself. Consequently every one knows things not as

they are, but as they are in the moment of perception for him, and

for him only ; and they are in this moment with reference to him

such as he represents them to himself. This is the meaning of the

Protagorean relativism, according to which things are for every

individual such as they appear to him ; and this he expressed in the

famous proposition that man is the measure of all things.

According to this, therefore, every opinion which grows out of per-

ception is true, and yet in a certain sense, just for this reason, it is

1 Whether these two motions were already designated by Protagoras as active

and passive (ttolovv and irdaxov), as is the case in Plato's presentation (Thecet.

156 A), may remain undecided. At all events, such anthropological categories in

the mouth of the Sophist are not surprising.2 With regard to such preparatory ideas, there is no ground to trace this

theory of the motions which advance to meet one another, to direct connectionwith Ileraclitus. Its Heraclitean element, which Plato very correctly saw, wassufficiently maintained by those direct predecessors who reduced all Becomingand change to relations of motion.

Page 113: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 8.] Problem of Science : Protagoras. 93

also false. It is valid only for the one perceiving, and for him even

only at the moment when it arises. All universal validity forsakes

it. And since, according to the view of Protagoras, there is no

other kind of ideas, and therefore no other knowledge than percep-

tion, there is for human knowledge nothing whatever that is univer-

sally vaiid. This view is phenomenalism in so far as it teaches in

this entirely definite sense a knowledge of the phenomenon, limited

to the individual and to the moment ; it is, however, scepticism in so

far as it rejects all knowledge which transcends that.

How far Protagoras himself drew practical consequences from this

principle that every one's opinion is true for himself, we do not

know. Later Sophists concluded that, according to this, error would

not be possible ; everything, and again nothing, belongs to everything

as attribute. In particular they concluded that no actual contradic-

tion is possible ; for since every one talks about the content of his

perception, different assertions can never have the same object. At

all events, Protagoras refused to make any positive statement con-

cerning what is; he spoke not of the actual reality that moves,

but only of motion, and of the phenomena which it produces for

perception.

Moreover, the attempt was now made, whether by Protagoras him-

self, or by the Sophistic activity dependent upon him, to trace dif-

ferences in perception, and so also in the phenomenon, back to

differences in this motion. It was principally the velocity of the

motion which was considered in this connection, though the form also

was probably regarded. 1 It is interesting to note further that under

the concept of perception not only sensations and perceptions, but

also the sensuous feelings and desires, were subsumed ; it is note-

worthy especially because to these states also an alaorjrov, a momen-tary qualification of the thing which produced the perception, was

held to correspond. The predicates of agreeableness and desir-

ability receive in this way the same valuation epistemologically

as do the predicates of sensuous qualification. What appears

agreeable, useful, and desirable to any one is agreeable, useful,

and desirable for him. The individual state of consciousness is

here, too, the measure of things, and no other universally valid

determination of the worth of things exists. In this direction

the Hedonism of Aristippus was developed out of the Protagorean

doctrine ; we know, teaches Aristippus, not things, but only their

1 Doubtless we have here asserting itself the development of the Pythagoreantheory of knowledge out of the Atomistic school, to which this reduction of the

qualitative to the quantitative was essential (cf. above, § 5), even though the So-phist declined from principle to enter into such metaphysical theories as Atomism.

Page 114: Windelband, History of Philosophy, Vol. 1

94 The Crreeks : Anthropologioa Period. [Part x.

worth for us, and the states {-rraO-q) into which they put us. These,

however, are rest and. indifference, violent motion and pain, or gentle

motion and pleasure. Of these only the last is worth striving for

(cf. above, §7,9).

4. Thus all courses of Sophistic thought issued in giving up truth

as unattainable. Socrates, however, needed truth, and on this account

he believed that it was to be attained if it were honestly sought for.

Virtue is knowledge ; and since there must be virtue, there must be

knowledge also. Here for the first time in history the moral con-

sciousness appears with complete clearness as an epistemological

postulate. Because morality is not possible without knowledge,

there must be knowledge ; and if knowledge is not here and nowexistent, it must be striven for as the lover seeks for the possession

of the loved object. Science is the yearning, struggling love for

knowledge,— <£iA.oo-o<£ia, philosophy (cf. Plat. Symp. 203 E).

Out of this conviction grow all the peculiarities of the Socratic l

doctrine of science, 2 and in the first place the bounds within which

he held knowledge to be necessary and therefore possible. It is

only a knowledge of the relations of human life that is necessary

for the ethical life ; only for these is a knowing necessary, and

only for these is man's knowing faculty adequate. Hypotheses as

to metaphysics and the philosophy of Nature have nothing to do

with man's ethical task, and they are left unconsidered by Socrates,

so much the rather as he shared the view of the Sophists that it

was impossible to gain a sure knowledge concerning them. Science

is possible only as practical insight, as knowledge of the ethical

life.

This view was formulated still more sharply by the Sophistic

successors of Socrates under the influence of his eudsemonistic

principle. For both Cynics and Cyrenaics science had worth only

so far as it affords to man the right insight which serves to makehim happy. With Antisthenes and Diogenes science was prized

not in itself, but as a means for controlling the desires and for

knowing man's natural needs ; the Cyrenaics said the causes of

perception (to, TreirotrjKOTa ra 7rd0rj) are for us as much matters of

indifference as they are unknowable; knowledge which leads to

happiness has to do only with our states, which we know with

certainty. Indifference toward metaphysics and natural science

1 Cf. Fr. Schleiermacher, Ueber den Werth des Sokrates als Philosophen (Ges.W. III., Bd. 2, pp. 287 ff.).

2[ Wissenschaftslehre. Wissenschaft, "scientia," "science," has here both

its subjective and objective sense ; knowledge as mental act, and knowledge asa body of truth. Hence Wissenschaftslehre means both "doctrine of science,"i.e. science of knowledge, and "scientific doctrine" i.e. philosophy.— Tr.]

Page 115: Windelband, History of Philosophy, Vol. 1

Chap. 2, § S.] Problem of Science : Socrates. 95

is with Socrates, as with the Sophists, the result of employment

with the inner nature of man.

5. It will remain a noteworthy fact for all time that a man whoso narrowed for himself the intellectual horizon of scientific research

as did Socrates, should yet determine within this the essential

nature of science itself, in a manner so clear and so authoritative

for all the future. This achievement was due essentially to his

opposition to the relativism of the Sojihists,— an opposition that was

a matter both of instinct and of positive conviction. They taught

that there are only opinions (8d£ai) which hold good for individuals

with psycho-genetic necessity ; he, however, sought a knowledge that

should be authoritative for all in like manner. In contrast with

the change and multiplicity of individual ideas he demanded the

one and abiding which all should acknowledge. He sought the

logical " Nature " (<£vo-is) as others had sought the cosmological

or ethical "Nature" (cf. § 7, 1), and found it in the concept or

general notion. Here, too, the view propounded was rooted in the

demand, the theory in the postulate.

The ancient thinkers, also, had had a feeling that the rational

thinking to which the}r owed their knowledge was something essen-

tially other than the sensuous mode of apprehending the world in

vogue in everyday life, or than traditional opinion ; but they had

not been able to carry out this distinction in relative worth either

psychologically or logically. Socrates succeeded in this because

here, too, he defined the thing in question by the work which he

expected it to perform. The idea that is to be more than opinion,

that is to serve as knowledge for all, must be what is commonin all the particular ideas which have forced themselves upon

individuals in individual relations : subjective universal validity is

to be expected only for the objectively universal. Hence, if there is

to be knowledge, it is to be found only in that in which all par-

ticular ideas agree. This universal in the object-matter which

makes possible the subjective community of ideas is the concept

(Adyos), and science [scientific knowledge] is accordingly conceptional

thinking,— abstract thought. The universal validity which is

claimed for knowledge is only possible on condition that the

scientific concept brings out into relief the common element which

is contained in all individual perceptions and opinions.

Hence the goal of all scientific work is the determination of the

essential nature of conceptions, — definition. The aim of investiga-

tion is to establish ti eKaarov ecrj, what each thing is, and to come to

ideas of an abiding nature as over against changing opinions.

Page 116: Windelband, History of Philosophy, Vol. 1

96 The Greeks : Anthropological Period. [Part 1

This doctrine was in some measure prepared for by the investigations of the

Sophists concerning the meaning of words, synonyms, and etymological rela-

tions. In the latter respect, the hypotheses of the Sophists in the beginnings of

the philosophy of language (cf. Plato's Cratylus) extended to the auestion

whether a natural or only a conventional relation obtains between worüs andtheir meanings (4>vo-et. 7} Öevei). Prodicus, whom Socrates mentions with com-mendation, seems to have been specially successful in fixing the meanings of words.

Among the later Sophists the Socratic demand for fixed conceptions becameforthwith fused with the Eleatic metaphysics, and with its postulate of the iden-

tity of Being with itself. Euclid called virtue, or the good, the only Being : it

remains the same, changeless in itself, and only the names by which men call

it differ. Antisthenes, indeed, explained the concept by the definition that it

is this which determines the timeless Being of the thing

;

l but he conceived

this identity of the existent with itself, raised above all relations, in so bold amanner that he thought of every truly existing entity as capable of being defined

only through itself. Predication is impossible. There are none but analytic

judgments (cf. above, No. 1). Accordingly only the composite can have its

essential elements determined in conceptions ; the simple is not to be defined.2

There is, then, no possibility of understanding the simple by conceptions ;it can

only be exhibited in a sensuous presentation. The Cynics came thus from the

Socratic doctrine of the conception to a sensualism which recognised as simple

and original only that which can be grasped with the hands and seen with the

eyes, and this is the ground of their opposition to Plato.

6. The searching out of conceptions (for his purpose, indeed, only

efchical conceptions) was accordingly for Socrates the essence of

science, and this determined in the first place the outer form of his

philosophising. The conception was to be that which is valid for

all : it must then be found in common thinking. Socrates is neither

a solitary hypercritic nor an instructor who teaches ex cathedra, but

a man thirsting for the truth, as anxious to instruct himself as to

teach others. His philosophy is a philosophy of the dialogue; it

develops itself in conversation which he was ready to begin with

every one who would talk with him. 3 To the ethical conceptions

which he alone was seeking for, it was indeed easy to find access

from any object whatever of everyday business. The commonelement must be found in the mutual exchange of thoughts ; the

SiaAoyioyAos was the way to the Aoyos. But this " conversation

"

encountered many difficulties : the inertia of the customary mode

of thinking, the idle desire for innovation, and the paradoxical state-

ments which were characteristic of the Sophists, the pride belong-

ing to seeming knowledge and thoughtless imitation. Into such a

condition of things Socrates made his entrance by introducing him-

self as one eager to learn. By skilful questions he drew out the

views of others, disclosed the defects in these views with remorse-

less consistency, and finally led the Athenian, proud of his culture,

into the state of mind where he recognised that insight into one's

1 X670S effTiv b to ri r/v rj tern dtjXüjv : Diog. Laert. VI. 3.

2 Plat. Thecct. 202 B.3 This factor united with the influence of Zeno's dialectic to stamp upon the

succeeding philosophical literature the form of the dialogue.

Page 117: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 8.] Problem of Science : Socrates. 9?

own ignorance, is the beginning of all knowledge. Whoever stood

this test and still remained with him was taken into partnership

in a serious effort to determine, in common thinking, the essential

meaning of conceptions. Undertaking the direction of the conver-

sation, Socrates brought his companion step by step to unfold his

own thoughts in clearer, less contradictory statements, and so caused

him to bring to definite expression what was slumbering in him as

an imperfect presentiment. He called this his art of mental mid-

wifery, and that preparation for it his irony.

7. The maieutic method has, however, still another essential

meaning. In the process of conversation the common rational

quality comes to light, to which all parts are subject in spite of

their diverging opinions. The conception is not to be made, it is

to be found ; it is already there, it requires only to be delivered fromthe envelopes of individual experiences and opinions in which it

lies hidden. The procedure of the Socratic formation of conceptions

is, therefore, epagogic or indtictive : it leads to the generic concep-

tion by the comparison of particular views and individual sensuous

presentations; it decides every individual question by seeking to

press forward to determine a general conception. This is accom-

plished by bringing together analogous cases, and by searching

out allied relations. The general conception thus gained is then

employed to decide the special problem proposed, and this subordi-

nation of the particular under the general is thus worked out as the

fundamental relation of scientific knowledge.

The inductive method of procedure as employed by Socrates,

according to Xenophon and Plato, is, to be sure, still marked by a

childlike simplicity and imperfection. It lacks as yet caution in

generalisation and methodical circumspection in the formation of

conceptions. The need for the general is so lively that it satisfies

itself at once with hastily gathered material, and the conviction of

the determining validity of the conception is so strong that the

individual questions proposed are decided forthwith in accordance

with it. But however great the gaps may be in the arguments of

Socrates, the significance of these arguments is by no means lessened.

His doctrine of induction has its value not for methodology, but for

logic, and for the theory of knowledge. It fixes in a way that is

decisive for all the future that it is the task of science to strive to

establish general conceptions from comparison offacts.

8. While Socrates thus defined the essential nature of science as

conceptional thought, — thinking in conceptions, — he also fixed the

bounds xoithin lohich science can be employed : this task is, in his

opinion, to be fulfilled only within the domain of practical life.

Page 118: Windelband, History of Philosophy, Vol. 1

98 The Greeks : Anthropological Period. [Part 1.

Science is, as regards its form, the formation of conceptions, and as

regards its content ethics.

Meanwhile the whole mass of ideas concerning Nature and all the

connected questions and problems still persist, and though for the most

part they are a matter of indifference for the moral life, neverthe-

less they cannot be entirely put aside. But after Socrates renounced

the task of attaining insight into such questions through conceptions,

it was all the more possible for him to form an idea of the universe

that should satisfy his scientifically grounded ethical needs.

So it comes that Socrates puts aside, indeed, all natural science,

but at the same time professes a teleological view of Nature, which

admires the wisdom in the arrangement of the world, the adaptation

in things, 1 and which, where understanding ceases, trusts Providence

in faith. With this faith Socrates kept himself as near as possible

to the religious ideas of his people, and even spoke of a plurality of

gods, although he indeed inclined to the ethical monotheism which

was preparing in his time. But he did not come forward in such

matters as a reformer : he taught morality, and if he expounded his

own faith, he left that of others untouched.

Out of this faith, however, grew the conviction with which he

limited the rationalism of his ethics,— his confidence in the Saifionov.

The more he pressed toward clearness of conceptions and complete

knowledge of ethical relations, and the more true to himself he was

in this, the less could he hide from himself that man in his limita-

tion does not completely succeed in this task, that there are condi-

tions in which knowledge is not sufficient for certain decision, and

where feeling enters upon its rights. Under such conditions Soc-

rates believed that he heard within himself the daimonion, a coun-

selling and for the most part warning voice. He thought that in

this way the gods warned from evil in difficult cases, where his

knowledge ceased, the man who otherwise served them.

So the wise man of Athens set faith and feeling beside ethical

science.

1 It is not probable that Socrates experienced any strong influence fromAnaxagoras in this respect, for the latter's teleology relates to the harmony of

the stellar universe, not to human life, while the considerations which areascribed to Socrates, especially by Xenophon, make utility for man the standardfor admiration of the world. Much more closely related to Socratic faith arethe religious views of the great poets of Athens, especially the tragedians.

Page 119: Windelband, History of Philosophy, Vol. 1

CHAPTER III.

THE SYSTEMATIC PERIOD.

The third, completing period of Greek science harvested the fruit

of the two preceding developments. It appears essentially as a

reciprocal inter-penetration ofcosmological and anthropological bodies of

thought. This union appears in but a very slight degree as a neces-

sity found in the nature of the case, still less as a demand of the

time ; rather, it is m its essentials the work of great personalities

and of the peculiar direction taken by their knowledge.

The tendency of the time was rather toward a practical utilisa-

tion of science : it was in accord with this tendency when research

separated into special investigations on mechanical, physiological,

rhetorical, and political problems, and when scientific instruction

accommodated itself to the ideas of the ordinary man. Not only for

the mass of the people, but for scholars as well, general questions of

cosmology had lost the interest which in the beginning was directed

toward them, and the fact that they were sceptically abandoned

because of the Sophistic theory of knowledge is nowhere presented

in the form of renunciation or lamentation.

If, therefore, Greek philosophy turned with renewed force from

the investigation of human thinking and willing— researches with

which it had busied itself during the time of the Enlightenment—back to the great problems of metaphysics, and reached its greatest

height along this path, it owes this achievement to the personal

thirst for knowledge on the part of the three great men whobrought in this most valuable development of ancient thought, and

stand as its representatives,

Democritus, Plato, and Aristotle.

The creations of these three heroes of Greek thought differ from the

doctrines of all their predecessors by reason of their systematic char-

acter. Each of the three gave to the world an all-embracing system

of science complete in itself. Their teachings gained this character,

on the one hand, through the all-sidedness of their problems, and on

the other, through the conscious unity in their treatment of them.

While each of the earlier thinkers had seized upon but a limited

99

Page 120: Windelband, History of Philosophy, Vol. 1

100 The Philosophy of the Greeks. [Part I.

circle of questions, and in like manner had shown himself informed

only in certain departments of actual reality, while especially no

one had as yet shown interest in both physical and psychological

investigation, these three men directed their work in like measure

to the entire compass of scientific problems. They brought together

what experience and observation had won ; they examined and com-

pared the conceptions which had been formed from these, and they

brought that which up to this time had been isolated, into fruitful

union and relation. This all-sidedness of their scientific interest

appears in the compass and varied character of their literary activ-

ity, and the great amount of material elaborated is in part explained

only through the vigorous co-operation of their extended schools, in

which a division of labour in accordance with inclination and endow-

ment was allowed.

But this work thus shared in common did not result in a mass of

unrelated material. This was guarded against by the fact that each

of these three men undertook and conducted the working over of the

entire material of knowledge with a unity of purpose and method

derived from the principle which formed his fundamental thought.

This, indeed, led at more than one point to a one-sided conception,

and to a kind of violation of individual domains, and thereby to

the inter-weaving of problems in ways which do not stand criticism.

But on the other hand, just by means of the adjustment which must

take place in this process between the forms of cognition in differ-

ent departments of knowledge, the formation of metaphysical concep-

tions was so furthered, abstract thought was so refined and deepened,

that in the short time of scarcely two generations the typical out-

lines of three different conceptions of the world were worked out.

Thus the advantages and the disadvantages of philosophical system-

building appear in like measure in the case of these men of genius

who were the first founders of systems.

The systematizing of knowledge so that it should become an all-in-

clusive philosophical doctrine was achieved with increasing success

by Democritus, Plato, and Aristotle, and with the last first found

the form of an organic articulation of science into the individual

disciplines. With this Aristotle concluded the development of Greek

philosophy and inaugurated the age of the special sciences.

The course of this development was more particularly this : the

two opposing systems of Democritus and Plato arose from the

application to cosinological and metaphysical problems, of the prin-

ciples gained through the doctrines of the Sophists and of Socrates

;

from the attempt to reconcile these opposites proceeded the conclud-

ing doctrine of Aristotle.

Page 121: Windelband, History of Philosophy, Vol. 1

Chap. 3.] The Systematic Period. 101

The essential feature in the work of Democritus and Plato was

that they used the insight into the theory of knowledge, gained by

the philosophy of the Enlightenment, to ground metaphysics anew.

Their common dependence upon the doctrines of the cosmological

period and upon the Sophistic teaching, in particular upon the the-

ory of Protagoras, stamps upon the two doctrines a certain parallel-

ism and a partial relationship, — a relationship the more interesting,

the deeper the contrast between the two in other respects. This

contrast, however, is due to the fact that the Socratic teaching had

no effect upon Democritus, while its influence on Plato was decisive;

hence the ethical factor is as preponderant in the system of the latter

as it is unimportant in that of the former. Thus in parallel lines from

the same source developed the materialism of Democritus and the

idealism of Plato.

From this contrast is explained, too, the difference in their wont-

ing. The purely theoretical conception of science which prevails

with Democritus did not suit the age ; his school soon disappeared.

Plato, on the contrary, whose scientific teaching furnished at the

same time the basis for a principle of life, had the pleasure of form-

ing in the Academy an extensive and lasting school. But this school,

the so-called Older Academy, following the general tendency of the

time, soon ran out partly into special investigation, partly into pop-

ular moralising.

Out of it rose then the great form of Aristotle, the most influential

thinker that history has seen. The powerful concentration with

which he caused the entire content of thought in Greek science to

crystallise about the conception of development (ivreXtxaa) in order

to adjust the opposition discovered between his two great predeces-

sors, made him the philosophical teacher of the future, and his system

the most perfect expression of Greek thought.

Democritus of Abdera (about 400-360) was educated in the scientific asso-ciation of his home and by journeys lasting many years, led the life of a quiet,

unassuming investigator in his native city during the turmoil of the Sophisticperiod, and remained far from the noisy activity of Athens. He did not impartany special ability, political or otherwise, by his teaching, but was essentiallydisposed to theoretical thought, and particularly inclined to the investigation ofNature. With gigantic learning and comprehensive information he united greatclearness of abstract thought and apparently a strong inclination to simplify prob-lems schematically. The number of his works proves that he stood at the headof an extended school, of which some unimportant names are preserved, yetnothing is more characteristic of the way in which his age turned aside fromresearch that was not interesting to it than the indifference with which his sys-tem of the mechanical explanation of Nature was met. His doctrine was forcedinto the background for two thousand years by the teleological systems, andprolonged its existence only in the Epicurean school, while even there it was notunderstood.

Antiquity honoured Democritus as a great writer also, and for this reason thealmost complete loss of his works is all the more to be lamented, as aside from

Page 122: Windelband, History of Philosophy, Vol. 1

102 The Philosophy of tie Cfreeks. [Part L

the numerous titles only very unimportant and in part doubtful fragments are

extant. The most important writings seem to have been, theoretically, the Mifya?

and Mt/cpos 5tdi<ocr/j.os, vepi vov and irepl l8eQv;practically, irepl evdvp.iy)% and virodij-

Kai. W. Kahl (Diedenhofen, 1889) has begun to work through the sources

which had been collected by W. Burchard (Minden, 1830 and 1834) and Lort-

zing (Berlin, 1873). P. Natorp has edited the Ethics (Leips. 181)3).

Cf. P. Natorp, Forschungen zur Geschichte des Erkenntnissproblems im Alter-

thum (Berlin, 1884); G. Hart, Zur Seelen- und Erkenntnisslehre des Demokrit(Leips. 1886).

Plato of Athens (427-347), of distinguished family, had most successfully

assimilated the artistic and scientific culture of his time when the personality of

Socrates made so decisive an impression upon him that he abandoned his at-

tempts at poetry and devoted himself entirely to the society of the master. Hewas his truest and most intelligent, and yet at the same time his most indepen-

dent disciple. The execution of Socrates occasioned his acceptance of Euclid's

invitation to Megara ; then he journeyed to Cyrene and Egypt, returned for a

time to Athens, and here began to teach through his writings, and perhaps also

orally. About 390 we find him in Magna Grsecia and Sicily, where he becameconnected with the Pythagoreans and took part also in political action. This

brought him into serious danger at the court of the ruler of Syracuse, the elder

Dionysius, whom he sought to influence with the help of his friend Dion ; he

was delivered as prisoner of war to the Spartans and ransomed only by the help

of a friend. This attempt at practical politics in Sicily was twice repeated later

(367 and 361), but always with unfortunate results.

After the first Sicilian journey, he founded his school in the grove Akademos,and soon united about him a great number of prominent men for the purpose

of common scientific work. Yet the bond of this society was to be sought still

more in a friendship based upon community of ethical ideals. His teaching

activity at the beginning had, like that of Socrates, that character of a commonsearch for truth which finds expression in the dialogue. It was not until his

old age that it took on more the form of the didactic lecture.

This life finds its aesthetic and literary embodiment in Plato's works, 1 in whichthe process itself of philosophising is set forth with dramatic vividness andplastic portraiture of personalities and their views of life. As works of art, the

Symposium and the Phoedo are most successful ; the grandest impression of

the system, as a whole, is afforded by the Republic. With the exception of the

Apology of Socrates, the form is everywhere that of the dialogue. Yet the

artistic treatment suffers in Plato's old age, and the dialogue remains only as

the schematic setting of a lecture, as in the Timazus and the Laws. For the

most part, Socrates leads the conversation, and it is into his mouth that Plato

puts his own decision when he comes to one. Exceptions to this are not founduntil in the latest writings.

The mode of presentation is also on the whole more artistic than scientific. It

exhibits extreme vividness and plasticity of imagination in perfect language, butno strictness in separating problems or in methodical investigation. The con-

tents of any individual dialogue is to be designated only by the prominent sub-

ject of inquiry. Where abstract presentation is not possible or not in place

Plato takes to his aid the so-called myths, allegorical presentations which utilise

motives from fables and tales of the gods in free, poetic form.

The transmission of his works is only in part certain, and it is just as doubtful

in what order they originated and what relation they bear to one another.

The following are among the most important names of those who have workedover these questions since Schleiermacher in his translation (Berlin, 1804 ff.)

gave an impulse in that direction: J. Socher (Munich, 1820), C. Fr. Hermann

1 Translated into German by Hier. Müller, with introductions by K. Steinhart.

8 vols. Leips. 1850-1866. As ninth volume of the series Platon's Leben, byK. Steinhart. Leips. 1873. [English by Jowett, third ed. 5 vols. Oxford,

1893. ] Among more recent editions, in which the paging of that of Stephanus(Paris, 1578), employed in citations, is always repeated, are to be noted those

of J. Bekker (Berlin, 1816 f.), Stallbaum (Leips. 1850), Schneider andHirschig (Paris: Didot, 1846 ff.), M. Schanz (Leips. 1875 ff.).

Page 123: Windelband, History of Philosophy, Vol. 1

Chap. 3.] The Systematic Period. 103

(Heidelberg, 1839), E. Zeller (Tübingen, 1839), Fr. Suckow (Berlin, 1855),Fr. Suseinihl (Berlin, 1855-5(3), E. Munk (Berlin, 1886), Fr. Ueberweg (Vienna,1861), K. Schaarschmidt (Bonn, I860), H. Bonitz (Berlin, 1875), G. Teich-müller (Gotha, 1876 ; Leipsic, 1879; Breslau, 1881), A. Krolm (Halle, 1878), W.Ditteuberger (in Hermes, 1881), H. Siebeck (Freiburg i. B. 1889). [H. Jack-son in Jour. Phil., X., XL, and XIII.; Archer-Hind's editions of Phcedo andTimmis; reviewed critically by P. Shorey in Am. Jour. Piniol., IX. and X.]

[On Plato's philosophy, in addition to the above, W. Pater, Plato and Platon-ism (Bond, and N.Y. 1893) ; J. Martineau, in Types of Ethical Theory (Lond.and N.Y. 1886), also in Essays ; Art. Plato in Euc. Brit., by L. Campbell ; R. L.Nettleship, The Theory of Education in P.'s Bep., in Bellenicc ; J. S. Mill in

Essays and Discussions.]The writings which are considered genuinely Platonic are (a) youthful works,

which scarcely go beyond the Socratic standpoint : Apology, Crito, Euthyphro,Lysis, Laches (perhaps also Charmides, Hippias Minor, and Alcibiades, I.)

;

(ft) writings to establish his position with regard to the Sophistic doctrines:

Protagoras, Gorgias, Euthydemus, Cratylus, Ueno, 7 hecetetus ; (c) main worksintended to present his own doctrine : Phcedrus, Symposium, Phcedo, Philebus,and the Bepublic, whose working out, begun early and completed in successivestrata, as it were, extended into the last years of the Philosopher's life

;(d) the

writings of his old age : Timceus, the Laws, and the fragment of Critias. Amongthe doubtful writings the most important are the Sophist, Politicus, and Par-menides. These probably did not originate with Plato, but with men of his

school who were closely related with the Eleatic dialectic and eristic. The first

two are by the same author.Cf. H. v. Stein, Sieben Bücher zur Geschichte des Piatonismus (Göttingen,

1861 ff.); G. Grote, Plato and the Other Companions of Socrates (Lond. 1865);A. E. Chaignet, La vie et les ecrits de Platon (Paris, 1873); E. Heitz, (0. Müller1

s

Gesch. der griech. Lit., 2. Aufl., IL 2, 148-235).Plato's school is called the Academy, and the time of its development, which

reaches to the end of ancient thought, and which was aided by the continuedpossession of the academic grove and the gymnasium existing there, is usually

divided into three or five periods : (1) the Older Academy, Plato's most imme-diate circle of scholars and the succeeding generations, extending to about 260b.c.; (2) the Middle Academy, which took a sceptical direction, and in whichan older school of Arcesilaus and a younger school of Carneades (about 160) are

distinguished; (3) the New Academy, which with Philo of Larissa (about 100)

turned back to the old dogmatism, and with Antiochus of Ascalon (about twenty-five years later) turned into the paths of Eclecticism. Concerning the two (or

four) later forms cf. Part II. ch. 1. Later the Neo-Platonic school took posses-

sion of the Academy. Cf. Part II. ch. 2.

To the Older Academy belonged men of great erudition and honourable per-

sonality. The heads of the school were Speusippus, the nephew of Plato,

Xenocrates of Chalcedon, Polemo and Crates of Athens ; beside these,

Philip of Opus and Heracleidea from Pontic Heraclea are to be mentionedamong the older, and Crantor among the younger members. Less closely

related with the school were the astronomers Eudoxus of Cnidos and the

Pythagorean Archytas of Tarentum. R. Heinze, Xenocrates (Leips. 1892).

Aristotle of Stagira towers far above all his associates in the Academy(384-322). As son of a Macedonian physician, he brought with him an inclina-

tion toward medical and natural science, when, at eighteen years of age, he

entered the Academy, in which as literary supporter and also as teacher, at first

of rhetoric, he early played a comparatively independent part, without acting

contrary to a feeling of reverent subordination to the master, by so doing.

It was not until after Plato's death that he separated himself externally from the

Academy, visiting, with Xenocrates, his friend Hermias, the ruler of Atarneus andAssus in Mysia, whose relative Pythias he afterwards married. After an appar-

ently transient stay at Athens and Mitylene, he undertook, at the wish of Philip

of Macedon, the education of the latter's son Alexander, and conducted it for

about three years with the greatest results. After this, he lived for some years

in his native city, pursuing scientific studies with his friend Theophrastus, and

together with him, in the year 335, founded in Athens his own school, which

had its seat in the Lyceum, and (probably on account of its shady walks) was

called the Peripatetic School.

Page 124: Windelband, History of Philosophy, Vol. 1

104 The Greeks: Systematic Period. [Takt I.

After twelve years of the greatest activity, he left Athens on account ofpolitical disturbances and went to Chalcis, where he died in the following year,of a disease of the stomach. Cf. A. Stahr, Aristotelia, I. (Halle, 1830).

Of the results of the extraordinarily comprehensive literary activity of Aris-

totle only the smallest part, but the most important part from the point of viewof science, is extant. The dialogues published by himself, which in the eyes ofthe ancients placed him on a level with Plato as an author also, are lost with theexception of a few fragments, and so also are the great compilations which withthe aid of his scholars he prepared for the different branches of scientific knowl-edge. Only his scientific didactic writings, which were designed as text-booksto be made the foundation of lectures in the Lyceum, are extant. The plan ofexecution in his works varies greatly ; in many places there are only sketchynotes, in others complete elaborations ; there are also different revisions of thesame sketch, and it is probable that supplementary matter by different scholarshas been inserted in the gaps of the manuscripts. Since the first complete edi-

tion prepared in ancient times (as it appears, on the occasion of a new discoveryof original manuscripts) by Andronicus of Rhodes (60-50 b.c.) did not separatethese parts, many critical questions are still afloat concerning it.

Cf. A. Stahr, Aristotelia, II. (Leips. 1832); V. Rose (Berlin, 1854); H. Bonitz(Vienna, 1862 ft); J. Bernays (Berlin, 1863); E. Heitz (Leips. 1865 and in thesecond ed. of O. Miiller's Gesch. der griech. Lit., II. 2, 236-321); E. Vallien

(Vienna, 1870 ff.).

This text-book collection, 1 as it were, is arranged in the following manner

:

(a) Logical treatises : the Categories, on the Proposition, on Interpretation,

the Analytics, the Topics including the book on the Fallacies — brought togetherby the school as "Organon" ; (b) Theoretical Philosophy : Fundamental Science(Metaphysics), the Physics, the History of Animals, and the Psychology ; to thethree last are attached a number of separate treatises

;(c) Practical Philosophy:

the Ethics in the Nicomachean and Eudemian editions and the Politics (whichlikewise is not complete)

;(d) Poietical or Poetical Philosophy : the Bhetoric

and the Poetic.

Fr. Biese, Die Philosophie des Aristoteles (2 vols., Berlin, 1835-42); A.Rosmini-Serbati, Aristotele Exposto ed Esaminato (Torino, 1858); G. II. Lewes,Aristotle, a Chapter from the History of Science (Lond. 1864) ; G. Grote,Aristotle (published from his literary remains, Lond. 1872).

[Trans, of the Psychology by E. Wallace (Camb. 1882) ; of the Ethics, byPeters (Lond. 1881), Welldon (Lond. and N.Y.), Williams (Lond. 1876), Chase(Lond. 1877), Hatch (Lond. 1879); of the Poetics, by Wharton (Camb. 1883) ; ofthe Politics, by Welldon (Camb. 1888), Jowett (2 vols., Oxford, 1885-88) ; of

the Rhetoric, by Welldon (Lond. and N.Y. 1886) ; also tr. of all of the above andof the Metaphysics, Organon, and History of Animals in the Bonn Library.Editions of the Politics with valuable introduction by Newman (Oxford, 1887,2 vols.); of the Ethics, by A. Grant. Cf. also Art. in Enc. Brit., Aristotle byA. Giant; T. H. Green in Works; A. C. Bradley, ASs Theory of the State, in

Hellenica. E. Wallace, Outlines of ASs Phil, is convenient for the student.]

§ 9. Metaphysics grounded anew in Epistemology and Ethics.

The great systematisers of Greek science exercised a swift but

just criticism upon the Sophistic doctrine. They saw at once that

among the doctrines of the Sophists but a single one possessed the

worth of lasting validity and scientific fruitfulness— the perception

theory of Protagoras.

1 Of the newer editions, that of the Berlin Academy (J. Bekker, Branclis,

Rose, Usener, Bonitz), 5 vols., Berlin, 1831-70, is made the basis of citations.

The Parisian edition (Didot) is also to be noticed (Diibner, Bussemaker, Heitz)5 vols., Paris, 1848-74.

Page 125: Windelband, History of Philosophy, Vol. 1

Chap. 0, § 9.] The New Metaphysics

.

105

1. This, therefore, became the starting-point for Democritus and

for Plato ; and both adopted it in order to transcend it and attack

the consequences which the Sophist had drawn from it. Both admit

that perception, as being itself only a product of a natural process,

can be the knowledge of something only which likewise arises and

passes away as transitory product of the same natural process.

Perception then gives only opinion (8d£a) ; it teaches what appears

in and for human view (called voyaw in Democritus with a genuine

Sophistic mode of expression), not what truly or really (irefj with

Democritus, wtws with Plato) is.

For Protagoras, who regarded perception as the only source of

knowledge, there was consequently no knowledge of what is. That

he took the farther step of denying Being altogether and declaring

the objects of perception to be the sole reality, behind which there

is no Being to be sought for,— this "positivist" conclusion is not

to be demonstrated in his case : the doctrine of "nihilism" ("there

is no Being ") is expressly ascribed by tradition only to G-orgias.

If, nevertheless, from any grounds whatever, a universally valid

knowledge (yv-qori-q yvoifxrj with Democritus, imaTTJ/xr] with Plato) was

to be again set over against opinions, the sensualism of Protagoras

must be abandoned and the position of the old metaphysicians, whodistinguished thought (oWoia), as a higher and better knowledge,

from perception, must be taken again (cf. § 6). Thus Democritus

and Plato both in like manner transcend Protagoras by acknowledg-

ing the relativity of perception, and looking to "thought" again for

knowledge of what truly is. Both are outspoken rationalists. 1

2. This new metaphysical rationalism is yet distinguished from

the older rationalism of the cosmological period, not only by its

broader psychological basis, which it owed to the Protagorean

analysis of perception, but also in consequence of this, by another

valuation of perception itself from the standpoint of the theory of

knowledge. The earlier metaphysicians, where they could not tit

the contents of perception into their conceptional idea of the world,

had simply rejected them as deceit and illusion. Now this illusion

had been explained (by Protagoras), but in such a way that while

surrendering its universal validity the content of perception might

yet claim at least the value of a transient and relative reality.

This, in connection with the fact that scientific knowledge was

1 Cf. Sext. Emp. Adv. Math. VIII. 56. The doctrine of Democritus withregard to "genuine " knowledge i.s most shaiply formulated in Sext. Emp. Adv.Math. VII. 139. Plato's attack upon the Protagorean sensualism is found prin-

cipally in the Thewtrtus, his positive rationalistic attitude in the P/uedrus, Sym-posimn, Republic, and Phcedo.

Page 126: Windelband, History of Philosophy, Vol. 1

106 The Q-reeks : Systematic Period. [Part L

directed toward the abiding " true " Being, led to a division in the

conception of reality, and with this the fundamental need of explana-

tory thought came to clear, explicit consciousness,— a need which

unconsciously lay at the basis of the beginnings of science. To the

two kinds of knowledge— so Democritus and Plato taught— cor-

respond two different kinds of reality: to perception a changing,

relative, transient reality or actuality ; to thought a reality homo-

geneous, absolute and abiding. For the former Democritus seems

to have introduced the expression phenomena; Plato designates it

as the world of generation, yeVecrts : the other kind of reality Democ-

ritus calls ra irerj ovra ; Plato, to ovtws ov or ovaia [that which really

is, or essence].

In this way perception and opinion gain a correctness which is

analogous to that of scientific thought. Perception cognises chang-

ing reality as thought cognises abiding reality. To the two modes

of cognition correspond two domains of reality. 1

But between these two domains there exists for this reason the

same relation, as regards their respective values, as obtains between

the two kinds of cognition. By as much as thought, the universally

valid act of consciousness, is above perception, the knowledge valid

only for individuals and for the particular, by so much is the true

Being higher, purer, more primitive, raised above the lower actuality

of phenomena and the changing processes and events among them.

This relation was especially emphasised and carried out by Plato

for reasons hereafter to be unfolded. But it appears also with Democ-

ritus, not only in his theory of knowledge, but also in his ethics.

In this way the two metaphysicians agree with the result which

the Pythagoreans (cf. § 5, 7, and § 6, 1) had likewise won from

their premises, viz. the distinction of a higher and lower kind of

reality. Nevertheless, in the presence of this similarity we are not

to think of a dependence ; in nowise in the case of Democritus,

who was a complete stranger to the astronomical view of the Pythag-

oreans, and scarcely in the case of Plato, who indeed later adopted

the astronomical theory, but whose idea of the higher reality (the

doctrine of Ideas) has an entirely different content. The case

rather is that the common, fundamental motive which came from

the conception of Being propounded by Parmenides, led in these

three quite different forms to the division of the world into a

sphere of higher and one of lower reality.

3. The pragmatic parallelism in the motives of the two opposed

systems of Democritus and Plato reaches a step farther, although

1 Best formulated in Plat., Tim. 27 D ff., especially 29 C.

Page 127: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 9.] The New Metaphysics : Demoeritus, Plato. 107

but a short step. To the world of perception belong, without doubt,

the specihc qualities of the senses, for these disclose their relativity

in the fact that the same thing appears differently to different senses.

But after we have abstracted these qualities, that which remains as

an object for the knowledge of the truly actual, is primarily the

form which things have, and both thinkers designated as the true

essential nature of things the pure forms (I8eai).

But it almost seems as though here they had nothing in commonbut the name, striking as this fact is ; for if Demoeritus understood

by the iSecu, which he also called axviJ'aTa) n^s atom-forms, while

Plato understood by his iSeai or clSrj the conceptions corresponding

to logical species {Gattungsbegriffe), then the apparently like state-

ment that the truly existent consists in " forms " has a completely

different meaning in the two authors. For this reason we must

here, too, remain in doubt as to whether we should see a parallel

dependence upon Pythagoreanism, which, to be sure, had previously

found the essence of things in mathematical forms, and whose influ-

ence upon the two thinkers may be assumed without encountering

any difficulties in the assumption itself. At all events, however, if

a common suggestion was present, it led to quite different results in

the two systems before us, and though in both of them knowledge

of mathematical relations stands in very close relation to knowledge

of true reality, these relations are yet completely different with the

respective thinkers.

4. The relationship thus far unfolded between the two rational-

istic systems changes now suddenly to a sharp opposition as soon as

we consider the motives from which the two thinkers transcended

the Protagorean sensualism and relativism, and observe also the

consequences which result therefrom. Here the circumstance be-

comes of decisive importance, that Plato ivas the disciple of Socrates,

while Demoeritus experienced not even the slightest influence from

the great Athenian sage.

With Demoeritus the demand which drives him to transcend the

position of Protagoras grows solely out of his theoretical need and

develops according to his personal nature,— the demand, namely,

that there is a knowledge, and that this, if it is not to be found in

perception, must be sought for in thought ; the investigator of Nat-

ure believes, as against all the Sophistic teaching, in the possibility

of a theory that shall explain phenomena. Plato, on the contrary,

sets out with his postulate of the Socratic conception of virtue.

Virtue is to be gained only through right knowledge ; knowledge,

however, is cognition of the true Being : if, then, this is not to be

found in perception, it must be sought for through thought. For

Page 128: Windelband, History of Philosophy, Vol. 1

108 The Greeks : Systematic Period. [Part l

Plato philosophy grows, according to the Socratic principle, 1 out of

the ethical need. But while the Sophistic friends of Socrates were

endeavouring to give to the knowledge that constituted virtue some

object in the form of a general life-purpose, the good, pleasure, etc.,

Plato wins his metaphysical position with one stroke, by drawing

the inference that this knowledge in which virtue is to consist must

be the cognition of what is truly real, the ovaia,— as opposed to

opinions which relate to the relative. In his case the knowledge

in which virtue is to consist demands a metaphysics.

Here, then, the ways are already parting. Knowledge of the

truly real was for Democritus, as for the old metaphysicians,

essentially an idea of the unchangeably abiding Being, but an idea

by means of which it should be possible to understand the

derivative form of reality which is cognised in perception. His

rationalism amounted to an explanation of phenomena, to be gained

through thought; it was essentially theoretical rationalism. For

Plato, on the contrary, knowledge of the truly real had its ethical

purpose within itself; this knowledge was to constitute virtue, and

hence it had no other relation to the world given through per-

ception than that of sharply defining its limits. True Being has

for Democritus the theoretical value of explaining phenomena; for

Plato, the practical value of being the object of that knowledge

which constitutes virtue. His doctrine is, as regards its original

principle, essentially ethical rationalism.

Democritus, therefore, persevered in the work undertaken in the

school of Abdera, — the construction of a metaphysics of Nature.

With the help of the Sophistic psychology he developed Atomismto a comprehensive system. Like Leucippus, he regarded emptyspace and the atoms moving in it as the true reality. He then

attempted not only to explain from the motion of these atoms

all qualitative phenomena of the corporeal world as quantitative

phenomena, but also to explain from these motions all mental

activities, including that knowing activity which is directed

toward true Being. Thus he created the system of materialism.

Plato, however, was led to the entirely opposite result by his

attachment to the Socratic doctrine, which proved to be of decisive

importance for his conception of the essential nature of science.

5. Socrates had taught that knowledge consists in general concep-

tions. If, however, this knowledge, in contrast with opinions, was

to be knowledge of what truly, actually is, there must belong to the

content of these conceptions that higher Being, that true essential

1 tiet forth most clearly in the Meno, 96 ft

Page 129: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 10.] System of Materialism : Democritus. 100

reality which, it was held, could be grasped only by thought, in

contrast with perception. The "forms" of true reality, knowledge

of which constitutes virtue, are the species or class-concepts (Gattungs-

begriffe), e'lSy). With this consideration, the Platonic conception of

the " Idea " first gains its complete determination.

So understood, Plato's doctrine of Ideas presents itself as the

summit of Greek philosophy. In it are combined all the different

lines of thought which had been directed toward the physical, the

ethical, the logical first principle (apxv or <£wis). The Platonic

Idea, the species or class-concept, is firstly the abiding Being in the

change of phenomena; secondly, the object of knowledge in the

change of opinions ; thirdly, the true end in the change of desires.

But this ovaca, from the nature of its definition, is not to be found

within the sphere of what may be perceived, and everything cor-

poreal is capable of being perceived. The Ideas are then something

essentially different from the corporeal world. True reality is

incorporeal. The division in the conception of reality takes on

accordingly a fixed form ; the lower reality of natural processes or

generation (yeVecns), which forms the object of perception, is the

corporeal world ; the higher reality of Being, which thought knows,

is the incorporeal, the immaterial world, twos votjtos. Thus the

Platonic system becomes immaterialism, or, as we call it after the

meaning given by him to the word "Idea," Idealism.

6. In the Platonic system, accordingly, we find perhaps the most

extensive interweaving and complication of problems which history

has seen. The doctrine of Democritus, on the contrary, is ruled

throughout by the one interest of explaining Nature. Howeverrich the results which this latter doctrine might achieve for this

its proper end, — results which could be taken up again in a later,

similarly disposed condition of thought, and then first unfold their

whole fruitfulness,— at first the other doctrine must surpass this,

all the more in proportion as it satisfied all needs of the time and

united within itself the entire product of earlier thought. Morepoints of attack for immanent criticism are perhaps offered by the

Platonic system than by that of Democritus ; but for Greek thought

the latter was a relapse into the cosmology of the first period, andit was Plato's doctrine that must become the system of the future.

§ 10. The System of Materialism.

The systematic character of the doctrine of Democritus consists

in the way in which he carried through in all departments of his

work the fundamental thought, that scientific theory must so far

Page 130: Windelband, History of Philosophy, Vol. 1

110 The Grreeks : Systematic Period. [Part L

gain knowledge of the true reality, i.e. of the atoms and theii

motions in space, as to be able to explain from them the reality

which appears in phenomena, as this presents itself in perception.

There is every indication (even the titles of his books would showthis) that Democritus took up this task by means of investigations

covering the entire compass of the objects of experience, and in this

connection devoted himself with as great an interest to the psy-

chological as to the physical problems. So much the more must weregret that the greater part of his teachings has been lost, and

that what is preserved, in connection with accounts of others,

permits only a hypothetical reconstruction of the main conceptions

of his great work, a reconstruction which must always remain

defective and uncertain.

1. It must be assumed in the first place that Democritus was

fully conscious of this task of science, viz. that of explaining the

world of experience through conceptions of the true reality. That

which the Atomists regard as the Existent, viz. space and the par-

ticles whirring in it, has no value except for theoretical purposes.

It is only thought in order to make intelligible what is perceived

;

but for this reason the problem is so to think the truly real that

it may explain the real which appears in phenomena, that at the

same time this latter reality may "remain preserved" 1 as some-

thing that " is " in a derived sense, and that the truth which inheres

in it may remain recognised. Hence Democritus knew very well

that thought also must seek the truth in perception, and win it out

of perception.2 His rationalism is far removed from being in con-

tradiction with experience, or even from being strange to experience.

Thought has to infer from perception that by means of which the

latter is explained. The motive which lay at the foundation of

the mediating attempts following the Eleatic paradox of acosmism

became with Democritus the clearly recognised principle of meta-

physics and natural science. Yet unfortunately nothing is nowknown as to how he carried out in detail the methodical relation

between the two modes of cognition, and how the process by which

knowledge grows out of perception in the particular instance was

thought by him.

More particularly, the theoretical explanation which Democritus

1 The very happy expression for this is dia<rd>£eiv ra <j>aivbfieva. Cf. also Arist.

Gen. et Corr. I. 832, 5 a.2 Hence, the expressions in which he recognised the truth in the phenome-

non ; e.g. Arist. De An. I. 2, 404 a 27, and the like. To attempt, however, to

construe out of this a " sensualism" of Democritus, as has been attempted byE. Johnson (Plauen, 1868), contradicts completely the accounts with regard to

his attitude toward Protagoras.

Page 131: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 10.] System of Materialism : Democritus. Ill

gave for the contents of perception consists, as with Leucippus, in

the reduction of all phenomena to the mechanics of atoms. Whatappears in perception as qualitatively determined, and also as in-

volved in qualitative change (aXXoiovfiefiov) , exists "in truth" only

as a quantitative relation of the atoms, of their order, and their

motion. The task of science is then to reduce all qualitative to

quantitative relatioris, and to show in detail what quantitative rela-

tions of the absolute reality produce the qualitative characteristics

of the reality which appears in phenomena. Thus, the prejudice in

favour of what may be perceived or imaged {anschaulich), as if spatial

form and motion were something simpler, more comprehensible in

themselves, and less of a problem than qualitative character and

alteration, is made the principle for the theoretical explanation of

the world.

Since this principle is applied with complete systematic rigour

to the whole of experience, Atomism regards the psychical life with

all its essential elements and values as also a phenomenon, and the

form and motion of the atoms which constitute the true Being of

this phenomenon must be stated by the explanatory theory. Thus

matter in its form and motion is regarded as that which alone is

truly real, and the entire mental or spiritual life as the derived,

phenomenal reality. With this the system of Democritus first

assumes the character of conscious, outspoken materialism.

2. In the properly physical doctrines, the teaching of Democritus

presents, therefore, no change in principle as compared with that of

Leucippus, though there is a great enrichment by careful detailed

investigation. He emphasised still more sharply than his predeces-

sor, where possible, the thought of the mechanical necessity (avdyn-q^

which he also occasionally called Aoyos), in accordance with which

all occurrence or change whatever takes place, and further defined

this thought as involving that no operation of atoms upon one

another is possible except through impact, through immediate con-

tact, and further, that this operation consists only in the change of

the state of motion of the atoms which are also unchangeable as

regards their form.

The atom itself as that which "is," in the proper sense of the

word, has accordingly only the characteristics of abstract corpore-

ality, viz. the filling of a limited space, and the quality of being

in motion in the void. Although all are imperceptibly small, they

yet exhibit an endless variety of forms (i8«u or a-xv^ara). To form,

which constitutes the proper fundamental difference in the atoms,

belongs in a certain sense also size;yet it is to be observed that

the same stereometrical form, e.g. the sphere, may appear in different

Page 132: Windelband, History of Philosophy, Vol. 1

112 The G-reeks : Systematic Period. [Part J,

sizes. The larger the atom, the greater its mass ; for the essential

quality of what is, is indeed materiality, space-claiming. For this

reason Democritus asserted weight or lightness to be a function of

size, 1 evidently yielding to the mechanical analogies of daily life.

In connection with these terms (ßapv and kov<j>ov), however, we are

not to think of the falling motion, but solely of the degree of mechani-

cal movability or of inertia.2 Hence it was also his opinion that as

the atom-complexes whirled about, the lighter parts were forced out-

ward, while the more inert with their inferior mobility were gath-

ered in the middle.

The same properties communicate themselves as metaphysical

qualities to things which are composed of atoms. The form and

size of things is produced by the simple summation of the form and

size of the component atoms ; though in this case, the inertia is not

dependent solely upon the sum total of the magnitudes of the atoms,

but upon the greater or less amount of empty space that remains

between the individual particles when they are grouped together.

The inertia depends therefore upon the less or greater degree of

density. And since the ease with which particles may be displaced

with reference to one another depends upon this interruption of the

mass by empty space, the properties of hardness and softness belong

also to the true reality that is known by thought.

All other properties, however, belong to things not in them-

selves, but only in so far as motions proceeding from things act

upon the organs of perception ; they are " states of perception as it

is in process of qualitative change." But these states are also

conditioned throughout by the things in which the perceived prop-

erties appear, and here the arrangement and the situation which the

atoms have taken with reference to each other in the process of

composition are of principal importance. 3

While, then, form, size, inertia, density, and hardness are properties

of things irefj, i.e. in truth, all that is perceived in them by the indi-

vidual senses as colour, sound, smell, taste, exists only vo/iwor 0eW,

i.e. in the phenomenon. This doctrine, when taken up anew in the

philosophy of the Renaissance (cf. Part IV. ch. 2) and later, was

1 As the most extensive exposition for this and for the following topic The-ophr. De Sens. 61 ff. (Dox. D. 516) is to be compared.

2 It is scarcely to be decided now whether the motion of their own, whichAtomism ascribed to all the atoms as primitive and causeless, was thought oi

by Democritus as conditioned already by the size or mass, so that the greater

had, even from the beginning, possessed less velocity. At all events, these

determinations held good for him within the sphere of the mechanical operation

of the atoms on one another. What is larger can be pushed witb greater diffi-

culty ; what is smaller can be pushed more easily.3 Cf. Arist. Gen. et Corr. I. 2, 315 b 6.

Page 133: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 10.] System of Materialism : Democritus. 113

designated as distinguishing between the primary and secondary

qualities of things, and it is desirable to introduce this expression

here, since it corresponds throughout to the metaphysical and episte-

mological sense in which Democritus made the Protagorean doctrine

useful for his own purpose. While the Sophist would make all

properties secondary and relative, Democritus admitted this only for

the qualities perceived by special senses, and set over against these

the quantitative determinations as primary and absolute. He there-

fore designated also as " genuine knowledge " the insight into the

primary qualities to be won through thought, while, on the contrary,

perception which is directed toward the secondary qualities he

termed " obscure knowledge " (yv-qvi-q— ctkotlt] yvufx-rj).

3. The secondary qualities appear accordingly as dependent

upon the primary ; they are not, however, dependent upon these

alone, but rather upon the action of these upon the percipient

agent. But in the atomistic system that which perceives, the mindor soid, can consist only of atoms. To be more explicit, it consists,

according to Democritus, of the same atoms which constitute also

the essence of fire : namely, the finest, smoothest, and most mobile.

These are indeed scattered also through the whole world, and in so

far animals, plants, and other things may be regarded as animate, as

having souls, but they are united in largest numbers in the humanbody, where in life a fire-atom is placed between every two atoms of

other sorts, and where they are held together by breathing.

Upon this presupposition, then, analogous, as we see, to the older

systems, Democritus built up his explanation of phenomena from

the true essence of things. That is, perception, and with it the

secondary qualities, arises from the action of things upon the fire-

atoms of the soul. The reality which appears is a necessary result

of the true reality.

In carrying out this doctrine Democritus took up and refined the

theories of perception advanced by his predecessors. The effluxes

(cf. above, § 6, 3) which proceed from things to set in motion the

organs and through them the fire-atoms, he called images (ei'SwAa),

and regarded them as infinitely small copies of the things. Their

impression upon the fire-atoms is perception, and the similarity

between the content of this perception and its object was held to be

secured thereby. Since impact and pressure are the essence of all

the mechanics of the atoms, touch is regarded as the most primitive

sense. The special organs, on the contrary, were regarded as capable

of receiving only such images as corresponded to their own forma-

tion and motion, and this theory of the specific energy of the sense

organs was worked out very acutely by Democritus. From this it

Page 134: Windelband, History of Philosophy, Vol. 1

114 The Greeks : Systematic Period. [Part I.

followed also that in case there were things whose effluxes could

not act upon any one of the organs, these would remain imperceptible

for the ordinary man, and for these perhaps " other senses " might

be accessible.

This theory of images appeared very plausible to ancient thought.

It brought to definite expression, and indeed to a certain extent

explained, the mode of representing things which is still commonfor the ordinary consciousness, as if our perceptions were " copies "

of things existing outside of us. If one did not ask further howthings should come to send out such miniature likenesses of them-

selves into the world, he might think that he understood, by means

of this theory, how our " impressions " can resemble things with-

out. For this reason this theory at once attained the predominance

in physiological psychology, and retained its position until after the

beginnings of modern philosophy, where it was defended by Locke.

Its significance, however, for the conceptions in the system of

Democritus, lies in this, that it was regarded as describing that

motion of the atoms in which perception consists. It remained

hidden from this materialism, which was such from principle, as

well as from all its later transformations, that perception as a

psychical activity is something specifically different from any and

every motion of atoms, however determined. But in seeking out

the individual forms of motion from which the individual percep-

tions of the special senses arise, the philosopher of Abdera caused

many a keen observation, many a fine suggestion, to become known.

4. It is interesting now that the same fate befell the materialistic

psychology of Democritus as liad befallen the pre-Sophistic meta-

physicians (cf. § 6) : it, too, was obliged in a certain respect to oblit-

erate again the epistemologica2 contrast between perception and

thought. Since, that is, all psychical life is regarded as motion of

the fire-atoms, 1 and since the motion of atoms in the connected sys-

tem of the universe is conditioned by contact and impact, it follows

that thought, which knows the truly real, can be explained only from

an impression which this truly real makes upon the fiery atoms,—explained therefore itself only through the efflux of such images.

As a psychological process, therefore, thought is the same as percep-

tion, viz. impression of images upon fire-atoms ; the only difference

is that in the case of perception the relatively coarse images of the

atom-complexes are active, while thought, which apprehends true

reality, rests upon a contact of the fire-atoms with the finest images,

with those which represent the atomic structure of things.

1 Arist. De An. I. 2, 405 a 8.

Page 135: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 10.] System of Materialism : Democritus. 115

Odd and fantastic as this sounds, the indications are yet all in

favour of the supposition that Democritus drew this conclusion from

the presuppositions of his in iterialistic psychology. This psychol-

ogy knew no independent, internal mechanism of ideas or conscious

states, but only an arising of ideas through the motion of atoms.

Hence it regarded ideas that were evidently deceptive as also

" impressions," and sought for these the exciting images. Dreams,

e.g. were traced back to ei'SwAa which had either penetrated into the

body in the waking state and on account of their weak motion had

previously produced no impression, or had first reached the fiery

atoms in sleep, evading the senses. A mysterious ("magnetic," or

"psychic," we should say to-day) action of men upon one another

appeared comprehensible on this hypothesis, and an objective basis

was given to faith in gods and demons by assuming giant forms in

infinite space from which corresponding images proceeded.

In correspondence with this Democritus seems to have thought of

" genuine knowledge " as that motion of the fire-atoms which is pro-

duced by the impression of the smallest and finest images,— those

which represent the atomic composition of things. This motion is,

however, the most delicate, the finest, the gentlest of all— that which

comes nearest to rest. With this definition the contrast beticeen per-

ception and thought was expressed in quantitative terms— quite in the

spirit of the system. The coarse images of things as wholes set the

fiery atoms into relatively violent motion and produce by this means

the " obscure insight " which presents itself as perception ; the finest

images, on the contrary, impress upon the fiery atoms a gentle, fine

motion which evokes the " genuine insight " into the atomic structure

of things, i.e. thought. In consideration of this, Democritus com-

mends the thinker to turn away from the world of the senses, quite

in contrast with the mode of thought which would develop truth out

of perception. Those finest motions assert their influence only where

the coarser are kept back; and where too violent motions of the

fiery atoms take place, the result is false ideation, the aXko^povdv. 1

5. This same quantitative contrast of strong and soft, violent

and gentle motion, was laid by Democritus at the basis of his ethical

theory also. 2 In so doing he stood with his psychology completely

upon the wtellectualistic standpoint of Socrates in so far as he

transposed the epistemological values of ideas immediately into

ethical values of states of will. As from perception only that

i Theophr. Do Sens. 58 (Box. D. 515).2 The resemblance with the theory of Aristippus (§ 7, Q

) is so striking, that

the assumption of a causal connection is scarcely to be avoided. Yet it may bethat we should seek for this rather in a common dependence upon Protagoras,

than in the interaction of Atomism and Hedonism upon each other.

Page 136: Windelband, History of Philosophy, Vol. 1

116 The Grreeks : Systematic Period. [Part 1

obscure insight follows which, has for its object the phenomenon

and not the true essence, so also the pleasure which arises from the

excitation of the senses is only relative (i/opu>), obscure, uncertain

of itself, and deceitful. The true happiness, on the contrary, for

which the wise man lives "according to nature" (4>vo-u), the evSaipo-

vt'a, which is the end (i-eXo?) and measure (ot!pos) of human life, must

not be sought in external goods, in sensuous satisfaction, but only

in that gentle motion, that tranquil frame (cieo-rw), which attends

upon right insight, upon the gentle movement of the fiery atoms.

This insight alone gives to the soul measure and harmony (£upp,e-

Tpta), guards it from emotional astonishment (adav/xaaria), lends it

security and imperturbability (arapaiia, dOa/j.ßia) ,— the ocean-calm

(yaXyvrj) of the soul that has become master of its passions through

knowledge. True happiness is rest (fjo-vxia), and rest is secured only

by knowledge. Thus Democritus gains as the cap-stone of his

system his personal ideal of life, — that of pure knowledge, free

from all wishes ; with this ideal, this systematic materialism cul-

minates in a noble and lofty theory of life. And yet there is in it

also a tendency which characterises the morals of the age of the

Enlightenment : this peace of mind resting upon knowledge is the

happiness of an individual life, and where the ethical teachings of

Democritus extend beyond the individual, it is friendship, the rela-

tion of individual personalities to one another, that he praises,

while he remains indifferent as regards connection with the state

§ 11. The System of Idealism.

The origin and development of the Platonic doctrine of Ideas is

one of the most difficult and involved, as well as one of the most

effective and fruitful, processes in the entire history of European

thought, and the task of apprehending it properly is made still

more difficult by the literary form in which it has been transmitted.

The Platonic dialogues show the philosophy of their author in

process of constant re-shaping : their composition extended through

half a century. Since, however, the order in which the individual

dialogues arose has not been transmitted to us and cannot be estab-

lished absolutely from external characteristics, pragmatic hypotheses

based on the logical connections of thought must be called to our aid.

1. In the first place there is no question that the opposition

between Socrates and the Sophists formed the starting-point for

Platonic thought. Plato's first writings were dedicated to an

affectionate and in the main, certainly, a faithful presentation of

the Socratic doctrine of virtue. To this he attached a polemic

Page 137: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism: Plato. 117

against the Sophistic doctrines of society and knowledge marked by

increasing keenness, but also by an increasing tendency toward

establishing his own view upon an independent basis. The Platonic

criticism of the Sophistic theories, however, proceeded essentially

from the Socratic postulate. It admitted fully, in the spirit of

Protagoras, the relativity of all knowledge gained through percep-

tion, but it found just in this the inadequacy of the Sophistic theory

for a true science of ethics. 1 The knowledge which is necessary for

virtue cannot consist in opinions as they arise from the changing

states of motion in subject and object, nor can it consist of a

rational consideration and legitimation of such opinions gained by

perception

;

2it must have a wholly different source and wholly

different objects. Of the corporeal world and its changing states—Plato held to this view of Protagoras in its entirety— there is no

science, but only perceptions and opinions ; it is accordingly an

incorporeal world that forms the object of science, and this world

must exist side by side with the corporeal world as independently

as does knowledge side by side with opinion. 3

Here we have for the first time the claim of an immaterial reality,

brought forward expressly and with full consciousness, and it is

clear that this springs from the ethical need for a knowledge that

is raised above all ideas gained by sense-perception. The assump-

tion of immateriality did not at first have as its aim, for Plato, the

explanation of phenomena : its end was rather to assure an object

for ethical knowledge. The idealistic metaphysics, therefore, in its

first draft 4 builds entirely upon a new foundation of its own, with-

out any reference to the work of earlier science that had been

directed toward investigating and understanding phenomena ; it is

an immaterial Eleatism, which seeks true Being in the Ideas, with-

out troubling itself about the world of generation and occurrence,

which it leaves to perception and opinion.5

To avoid numerous misunderstandings 6 we must, nevertheless,

expressly point out that the Platonic conception of immateriality

(aaw/MiTov) is in nowise coincident with that of the spiritual or

psychical, as might be easily assumed from the modern mode of

thinking. For the Platonic conception the particular psychical

1 On this point, the Thecetetus brings together the whole criticism of the

Sophistic doctrine.2 86£a ixXtjötjs nerä \6you, Thecet. 201 E. (Probably a theory of Antisthenes.)3 Arist. Met. I. 0, 987 a 32 ; XIII. 4, 1078 b 12.4 As set forth in the dialogues Phcedrus and the Symposium.5 Investigations as to theoretical and natural science are first found in the

latest dialogues.6 To which the Neo-Pythagorean and Neo-PIatonic transformation of the

doctrine of Ideas gave occasion. Cf. Pt. II. ch. 2, § 18.

Page 138: Windelband, History of Philosophy, Vol. 1

318 The Greeks: Systematic Period. [Part I.

functions belong to the world of Becoming, precisely as do those of

the body and of other corporeal things ; and on the other hand, in

the true reality the " forms " or " shapes " of corporeality, the Ideas

of sensuous qualities and relations, find a place precisely as do those

of the spiritual relations. The identification of spirit or mind and

incorporeality, the division of the world into mind and matter, is un-

Platonic. The incorporeal world which Plato teaches is not yet the

spiritual.

Rather, the Ideas are, for Plato, that incorporeal Being which is

knoivn through conceptions. Since, that is, the conceptions in which

Socrates found the essence of science are not given as such in the

reality that can be perceived, they must form a " second," " other "

reality, different from the former, existing by itself, and this imma-

terial reality is related to the material, as Being to Becoming, as the

abiding to the changing, as the simple to the manifold— in short,

as the world of Parmenides to that of Heraclitus. The object of

ethical knowledge, cognised through general conceptions, is that

which " is " in the true sense : the ethical, the logical, and the phys-

ical dpxv (ground or first principle) are the same. This is the point

in which all lines of earlier philosophy converge.

2. If the Ideas are to be " something other " than the percep-

tible world, knowledge of them through conceptions cannot be found

in the content of perception, for they cannot be contained in it.

With this turn of thought, which corresponds to the sharper separa-

tion of the two worlds, the Platonic doctrine of knowledge becomes

much more rationalistic than that of Democritus, and goes also

decidedly beyond that of Socrates ; for while the latter had devel-

oped the universal out of the opinions and perceptions of individuals

inductively, and had found it as the common content in these opin-

ions and perceptions, Plato does not conceive of the process of

induction in this analytical manner, but sees in perceptions only the

suggestions or promptings with the help of which the soul bethinks

itself of the conceptions, of the knowledge of the Ideas.

Plato expressed this rationalistic principle in the form that phil-

osophical knowledge is recollection (dva/AVT/crts). He showed in the

example of the Pythagorean proposition l that mathematical knowl-

edge is not extracted from sense-perception, but that sense-percep-

tion offers only the opportunity on occasion of which the soul

recollects the knowledge already present within her, that is, knowl-

edge that has purely rational validity. He points out that the pure

mathematical relations are not present in corporeal reality ; on the

1 Mem, 80 ff.

Page 139: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism : Plato. 119

contrary, the notion of these relations arises in us when similar

figures of perception offer but the occasion therefor, and he extended

this observation, which is completely applicable to mathematical

knowledge, to the sum total of scientific knowledge.

That this reflection upon what is rationally necessary should be

conceived of as recollection is connected with the fact that Plato,

as little as any of his predecessors, recognises a creative activity of

the consciousness, which produces its content. This is a general

limit for all Greek psychology ; the content for ideas must somehowbe given to the " soul "

;hence, if the Ideas are not given in perception,

and the soul nevertheless finds them in herself on occasion of per-

ception, she must have already received these Ideas in some way or

other. For this act of reception, however, Plato finds only the

mythical representation, 1 that before the earthly life the souls have

beheld the pure forms of reality in the incorporeal world itself, that

the perception of similar corporeal things calls the remembrance

back to those forms forgotten in the corporeal earthly life, and that

from this awakes the philosophical impulse, the love of the Ideas

(Ipws), by which the soul becomes raised again to the knowledge

of that true reality. Here, too, as in the case of Democritus, it is

shown that the entire ancient rationalism could form no idea of

the process of thought except after the analogy of sensuous percep-

tion, particularly that of the sense of sight.

What Socrates in his doctrine of the formation of conceptions haddesignated as induction, became transformed, therefore, for Plato,

into an intuition that proceeds by recollecting (crvvayoryy), into re-

flection upon a higher and purer perception (Anschauung). This

pure perception, however, yields a plurality of ideas corresponding

to the multiplicity of objects which occasion such perceptions, and

from this grows the further task for science to know also the rela-

tions of the Ideas to each other. This is a second step of Plato's

beyond Socrates, and is specially important for the reason that it led

shortly to the apprehension of the logical relations between concep-

tions. It was principally the relations of the subordination and co-

ordination of concepts to which Plato became attentive. The

division of the class-concepts or logical genera into their species

played a great part in his teaching. 2 The possibility or impossibility

of the union of particular conceptions is brought more exactly into

1 Phwdr. 246 ff.

2 Cf. Phileb. 16 C. Yet this dividing process is not anywhere especially promi-nent in the writings that are certainly Platonic. It is handled with the pedantryof a school in the Sophist and Politicus. Antiquity preserved "definitions"and " divisions " from the Platonic school. In Athoicrus, II. ?<9 C, is an instance

of mockery, by a comic poet, at this academical concept-splitting.

Page 140: Windelband, History of Philosophy, Vol. 1

120 The G-reeks : Systematic Period. [Part L

consideration, 1 and as a methodical aid he recommended the hypo-

thetical method of discussion, which aims to examine a tentatively

proposed conception by developing all the possible consequences

that would follow from the possibility of its union with conceptions

already known.

These logical operations taken as a whole, by means of which the

Ideas and their relations to one another (kowdvuz) were to be found,

Plato denoted by the name dialectic. What is found in his writings

concerning it has throughout a methodological character, but is not

properly logical.

3. The doctrine of knowledge as recollection stood, however, in

closest connection with Plato's conception of the relation of Ideas to

the world ofphenomena. Between the higher world of ova-La and the

lower world of yei/ecns, between what is and what is in process of

Becoming, he found that relation of similarity which exists between

archetypes (TrapaSety/xara) and their copies or images (ei'ScoAa) . In this,

too, a strong influence of mathematics upon the Platonic philosophy

is disclosed : as the Pythagoreans had already designated things as

imitations of numbers, so Plato found that individual things always

correspond to their class-concepts only to a certain degree, and that

the class-concept is a logical ideal which none of its empirical

examples comes up to. He expressed this by the conception of

imitation (/At/ar/o-ts). It was thus at the same time established that

that second world, that of the incorporeal Ideas, was to be regarded

as the higher, the more valuable, the more primitive world.

Yet this mode of representing the matter gave rather a deter-

mination of their respective values than a view that was usable for

metaphysical consideration : hence Plato sought for still other desig-

nations of the relation. The logical side of the matter, according to

which the Idea as class-concept or species represents the total uni-

tary extent or compass, of which the individual things denote but a

part, appears in the expression participation (/xiOeiis), which means

that the individual thing but partakes in the universal essence of the

Idea ; and the changing process of this partaking is emphasised by

the conception of presence {-rrapova-La) . The class-concept or species

is present in the thing so long as the latter possesses the qualities

which dwell in the Idea. The Ideas come and go, and as these nowcommunicate themselves to things and now again withdraw, the

qualities in these things which are like the Ideas are successively

changed to the eye of perception.

The precise designation of this relation was, for Plato, an object

' Phcedo, 102 ff.

Page 141: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism : Plato. 121

of only secondary interest, provided only the difference between

the world of Ideas and the corporeal world, and the dependence of

the latter upon the former, were recognised. 1 Most important and

sufficient for him was the conviction that by means of conceptions

that knowledge which virtue needs of what truly and really is, could

be won.

A. Peipers, Ontologia Platonica. Leips. 1883.

4. But the logico-metaphysical interest which Plato grafted upon

the Socratic doctrine of knowledge carried him far beyond the

master as regards the contents of this doctrine. The general

characteristics which he developed for the essence of the Ideas

applied to all class-concepts, and the immaterial world was therefore

peopled with the archetypes of the entire world of experience. So

many class-concepts, so many Ideas ; for Plato, too, there are count-

less " forms." In so far criticism 2 was right in saying that Plato's

world of Ideas was the world of perception thought over again in

conception.

In fact, according to the first draft of the Platonic philosophy,

there are Ideas of everything possible, of things, qualities, and

relations ; of the good and the beautiful as well as of the bad and

the ugly. Since the Idea is defined methodologically, in a purely

formal way, as class-concept, every class-concept whatever belongs

to the higher world of pure forms ; and in the dialogue Parmenides,3

not only was Plato's attention called by a man schooled in the

Eleatic Sophistic doctrine to all kinds of dialectical difficulties

which inhere in the logical relation of the one Idea to its manycopies, but he was also rallied, spitefully enough, with the thought

of all the foul companions that would be met in his world of pure

conceptual forms.

Plato's philosophy had no principle that could serve as a weapon

against such an objection, nor is there in the dialogues any intima-

tion that he had attempted to announce a definite criterion for the

selection of those class-concepts that were to be regarded as Ideas,

as constituents of the higher incorporeal world. Nor do the ex-

amples which he adduces permit such a principle to be recognised

;

we can only say that it seems as if in course of time he continually

emphasised more strongly the attributes expressing worth (as the

good and the beautiful), the mathematical relations (greatness and

smallness, numerical determinations, etc.), and the types of species

in the organic world, while, on the contrary, he no longer reckoned

1 Phcedo, 100 D. 2 Arist. Met. I. 9, 990 b 1. 8 Parm. 130 C.

Page 142: Windelband, History of Philosophy, Vol. 1

122 The Greeks : Systematic Period. [Part 1

among the Ideas mere concepts of relation, especially negative

notions and things made by human art.1

5. Our knowledge of the systematic connection and order which

Plato intended to affirm in the realm of Ideas remains ultimately as

obscure as that in regard to the preceding point. Urgent as he was

to establish co-ordination and subordination among the conceptions,

the thought of a logically arranged pyramid of conceptions which

must culminate in the conception that was most general and poorest

in content seems not to have been carried out. A very problematical

attempt to set up a limited number (five) of most general concep-

tions 2is presented in the Sophist (254 ff.). But these attempts,

which tend toward the Aristotelian doctrine of the categories, are

not to be traced back with certainty to Plato himself.

With him we find, rather, only the doctrine presented in the

Philebus, as well as in the Republic, that the Idea of the Good is the

highest, embracing, ruling, and realising all others. Plato defines

this Idea as regards its content as little as did Socrates; he de-

termined it only by means of the relation, that it should represent

in its content the highest absolute end of all reality, of the incor-

poreal as of the corporeal. The subordination of the other Ideas

to this highest Idea is accordingly not the logical subordination of

a particular under the general, but the teleological of the means to

the end.

In the latest period of his philosophising, concerning which wehave only intimations in the Laws and in critical notices of Aris-

totle,3 and in the teachings of his nearest successors, the imperfec-

tion of this solution of the logical problem seems to have led Plato

to the unfortunate thought of developing the system of Ideas ac-

cording to the method of the Pythagorean number-theory. ThePythagoreans also, to be sure, had the purpose of attaching the

abiding arrangements of things symbolically to the development of

the number series. But that was only a makeshift, because they

had as yet no idea of the logical arrangement of conceptions : hence,

when Plato, in connection with his other thoughts, fell back upon

this makeshift, designated the Idea of the Good as the h, the One,

and attempted to derive from it the duality (Sua?) of the Infinite or

Indefinite, and the Measure (airupov and iripas, = even and odd ; cf.

§ 4, 11), and from this, further, the other Ideas in such a way as to

present a series of the conditioning and the conditioned, neither

i Cf. also Arist. Met. XII. 3, 1070 c 18.2 Being, rest, motion, sameness (tcivtöttjs) and otherness (erepdrris), i.e. the

division of Being into the resting (ova-La), ever the same -with itself, and the

moved (yeveais), in process of constant change.3 Cf. A Trendelenburg, Piatonis de Ideis et Numeris Doctrina (Leips. 1826).

Page 143: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism : Plato. 123

this deplorable construction nor the fact that men like Speusippus,

Xenocrates, Philippus, and Archytas undertook to carry it out in

detail, would be worth more particular mention, were not this just

the point to which the speculation of the ISTeo-Pythagoreans and the

Neo-Platonists became attached. For by this gradation which Plato

thus began within the ovala the world of true reality, the division

in the conception of reality, which had developed out of the opposi-

tion between perception and thought, became multiplied, and thus

dualism was again abolished. For when to the One, or the Idea of

the Good, was ascribed the highest absolute reality, and to the vari-

ous strata of the world of Ideas, a reality of constantly decreasing

worth in proportion as they were removed from the One in the

system in numbers, there arose from this a scale of realities which

extended from the One down to the lowest reality,— that of the

corporeal world. Fantastic as this thought may be, it yet evinced

its force and influence in the development of thought, even to the

threshold of modern philosophy. Its power, however, lies doubtless

in all cases in its amalgamation of attributes of worth with these

various grades of reality.

6. While as metaphysics, the doctrine of Ideas fell into such seri-

ous difficulties, it was carried out in an extremely happy, simple, and

transparent manner in that domain which formed its proper home,

— that of ethics. For the systematic elaboration of this, however,

Plato needed a psychology, and that, too, of another sort than the

psychology which had arisen in previous science, out of the presup-

positions of natural philosophy, and with the aid of individual per-

ceptions or opinions. When, in contrast with this, he developed

his psychology from the postulates of the doctrine of Ideas, the

result was of course a purely metaphysical theory which stood and

fell with its postulate, yet it was at the same time, by reason of the

import of the doctrine of Ideas, a first attempt to understand the

psychical life from within, and in accordance with its internal char-

acter and articulation.

The conception of the soid or mind was in itself a difficulty 1 in

the dualism of the doctrine of Ideas. For Plato, also, "soul" was

on the one hand the living element, that which is moved of itself

and moves other things, and on the other hand, that which perceives,

knows, and wills. As principle of life and of motion, the soul

belongs, therefore, to the lower world of Becoming, and in this it

remains when it perceives and directs its desires toward objects of

the senses. But this same soul, nevertheless, by its true knowledge

1 Fhcedu, 7G ff., 105, Phcedr. 245, Laws, X. 896.

Page 144: Windelband, History of Philosophy, Vol. 1

124 The Greeks: Systematic Period. [Pakt I.

of the Ideas, becomes partaker in the higher reality of abiding Being.

Hence it must be assigned a position betiveen the two worlds— not

the timeless, unchanged essence of the Ideas, but a vitality which

survives change ; i.e. immortality. Here, for the first time, personal

immortality is brought forward by Plato as a part of philosophic

teaching. Of the proofs which the Phmdo adduces for this, those

are most in accord with the spirit of the system which reason from

the soul's knowledge of Ideas to its relationship with eternity ; in

correspondence with the form of the system is the dialectic false

conclusion that the soul cannot be or become dead, because its

essential characteristic is life ; the most tenable of the arguments is

the reference to the unity and substantiality which the soul evinces

in ruling the body.

In consequence of this intermediate position the soul must bear in

itself the traits of both worlds ; there must be in its essence some-

thing which corresponds to the world of Ideas, and something

which corresponds to the world of perception. The former is the

rational nature (XoyiariKov or voSs), the seat of knowledge and of the

virtue which corresponds to it ; in the latter, the irrational nature,

Plato made a further distinction of two elements, — the nobler, which

inclines towards the Reason, and the lower, which resists it. Thenobler he found in the ardent, spirited Will (Spirit, 0v/xos), the

lower in the sensuous desire (Appetite, lindviAia.) . Thus Reason,

Spirit, and Appetite are the three forms of activity of the soul, the

classes or species (eiS??) of its states.

These fundamental psychological conceptions which had thus grownout of considerations of ethical worth are employed by Plato to set

forth the moral destiny of the individual. The fettering of the

soul to the body is at once a consequence and a punishment of

the sensuous appetite. Plato extends the immortal existence of

the soul equally beyond the two boundaries of the earthly life.

The sin for the sake of which the soul is ensnared in the world of

sense is to be sought in a pre-existent state

;

1its destiny in the

hereafter 1 will depend upon how far it has freed itself in the earthly

life from the sensuous appetite, and turned to its higher vocation—knowledge of the Ideas. But inasmuch as the ultimate goal of the

soul appears to be to strip off the sensuous nature, the three forms

of activity are designated also as parts of the soid. In the Timceus

Plato even portrays the process of the formation of the soul out of

these parts, and retains immortality for the rational part only.

1 These doctrines are depicted in the form of mythical allegories which makeuse of motives from the popular faith and from the Mystery-cults. V. Phcedr.246 ff.j Gorgias, 523 ff.; Hep. 614 ff.; Phcedo, 107 ff.

Page 145: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism : Plato. 125

It is already clear from these changing determinations that the

relation of these three fundamental forms of the psychical life to

the none too strongly emphasised unity of the soul's nature was not

clearly thought out; nor is it possible to give to these conceptions

formed from the ethical need the significance of purely psychologi-

cal distinctions, such as have since been made. 1

7. But at all events there followed in this way, from the doctrine

of the two worlds, a negative morals that would fly from the world,

and in which the withdrawal from the world of sense and the spir-

itualisation of life were praised as ideals of wisdom. It is not only

the Phtedo that breathes this earnest disposition in its portrayal of

the death of Socrates ; the same ethical theory prevails in such dia-

logues as the Gorgias, the Thecetetits. and, in part, the Republic.

But in Plato's own nature the heavy blood of the thinker wasassociated with the light heart-beat of the artist, and thus while his

philosophy lured him into the realm of bodiless forms, the whole

charm of Hellenic beauty was living and active within him.

Strongly as he therefore combated root and branch the theory of

Aristippus, which would fain regard man's strivings as satisfied

with sensuous pleasure, it was nevertheless his opinion that the

Idea of the Good becomes realised even in the world of sense.

Joy in the beautiful, pleasure in the sensuous imitation of the Idea,

painless because free from the element of wishing, the development

of knowledge and practical artistic skill, the intelligent understand-

ing of the mathematical relations which measure empirical reality,

and the appropriate ordering of the individual life, — all these were

valued by him as at least preparatory stages and participations in

that highest good which consists in knowledge of the Ideas, and of

the highest among them, the Idea of the Good. In the Symposiumand in the Philebus he has given expression to this his estimate of

the goods of life.

This same thought, that ethical values and standards must illu-

mine the whole circuit of human life, was used in another form by

Plato in that presentation of the system of the virtues which he

developed in the Republic. Here he showed that each part of the

soul has a definite task to fulfil, and so a perfection of its own to

reach : the rational part, in icisdom (o-o<f>ia), the spirited (ou/toeiSe's)

in energy of will (courage, avSpla), the appetitive (iniöviJirjTLKov) in

1 That the question here for Plato was essentially that of the gradation of thepsychical from the point of view of relative worth, is shown not only in theemployment made of these distinctions in ethics and politics, but also in suchremarks as those which designated this triple division as characteristic for thedifferent organic beings (plant, animal, man), or for the different peoples,

inhabitants of southern countries, of northern countries, and the Greeks.

Page 146: Windelband, History of Philosophy, Vol. 1

126 The Greeks : Systematic Period. [Part I.

self-control (moderation, awcppoavvr)) ; that, however, in addition to

all these, as the virtue of the soul as a whole, there must be the

right relation of these parts, complete uprightness (justice, BiKaioawy])

.

The true significance, however, of these four cardinal virtues, is

first unfolded upon a higher domain, that of politics.

8. The tendency of the doctrine of Ideas, directed as it was

toward the general and the universal, exhibited its most perfect

operation in the aspect now to be noticed, viz. that the ethical

ideal of the Platonic philosophy lay not in the ability and happi-

ness of the individual, but in the ethical perfection of the sjiecies.

True to the logical principle of the doctrine of Ideas, that which

truly is in the ethical sense, is not the individual man, but mankind,

and the form in which this truly existent humanity appears is the

organic union of individuals in the state. The ethical ideal becomes

for Plato the political, and in the midst of the time which saw the

dissolution of Greek political life, and in opposition to those doc-

trines which proclaimed only the principle of individual happiness,

he raised the conception of the state to an all-controlling height.

He considered the state, however, not from the side of its empiri-

cal origin, but in reference to its task, viz. that of presenting in

large the ideal of humanity, and of educating the citizen to that

particular virtue which makes him truly happy. Convinced that

his project could be realised, with force if necessary, he wove into

its fabric not only features which he approved of the then-existing

Greek political life, in particular those of the aristocratic Doric

constitutions, but also all the ideals for whose fulfilment he hoped

from the right formation of public life.

K. F. Hermann, Ges. Abhandlungen, 122 ff. ; E. Zeller, Vorträge und Ab-handlungen, I. 62 ff.

If the ideal state is to present man in large, it must consist of the

three parts which correspond to the three parts of the soul,— the

teaching class, the warrior class, and the working class. It belongs

to the first class alone, that of the cultured (<£iaoo-o<£oi), to guide the

state and to rule 1(äpxovre^), to give laws and to watch over their

observance. The virtue proper to this class is wisdom, insight into

that which is for the advantage of the whole, and which is demanded

by the ethical aim of the whole. To support this class there is the

second class, that of the public officials (eViKoupoi; guardians, <£vAaKes),

which has to evince the virtue of the fearless performance of duty

(dvSpta) as it maintains the order of the state within and without.

1 Hence the XoyurriKSv is called also -rjyeixoviKbv.

Page 147: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism : Plato. 127

It is, however, obedience which holds the desires in check, self-control

(auxfrpoavvr)) , that becomes the great mass of the people, the artisans

and farmers (yewpyoi kcu Sr/fxtovpyoi), who have to care for providing

for the external means of the states by their labour and industry. 1

Only when each class thus does its duty and maintains its appro-

priate virtue does the nature of the state correspond to the ideal of

justice (SiKaiocrwT/).

The principle of aristocracy in education, which is of decisive im-

portance in the Platonic ideal of the state, appears most clearly in

the provision that for the great mass of the third class only the

ordinary ability of practical life is claimed, and in that this is re-

garded as sufficient for their purpose, while the education, which the

state has the right and duty to take in hand itself in order to train

its citizens for its own ends, is given only to the two other classes.

By means of a constantly repeated process of selection continued

from birth to the late years, the government causes the two upper

classes to be continually renewed, strata by strata; and in order that

no individual interest may remain to hold back these classes, who are

properly the organs of the whole body, in the fulfilment of their

task, they are to renounce family life and private property. Their

lot is that of education by the state, absence of family relations,

community of life and of goods. He who is to live for the ends of the

whole, for the ethical education of the people, must not be bound to

the individual by any personal interest. To this thought, which

found its historic realisation in the sacerdotal state of the mediaeval

hierarchy, is limited whatever of communism, community of wives,

etc., men have professed to discover in the Platonic teaching. The

great Idealist carries out to its extreme consequences the thought

that the end of human life consists in moral education, and that

the entire organisation of a community must be arranged for this

sole end.

9. With this a new relation between the world of ideas and the

world of phenomena was discovered, and one which corresponded

most perfectly to the spirit of the Platonic system : the Idea of the

Good disclosed itself as the task, as the end (WAos), which the

phenomenon of human life in society has to fulfil. This discovery

became of decisive importance for the final form taken by Plato's

metaphysical system.

For, as first projected, the doctrine of Ideas had been precisely as

incompetent as the Eleatic doctrine of Being to explain empirical

reality. The class-concepts were held to give knowledge of the

1 Hence the third par* of the soul is called also the ^tXoxp^MaTov.

Page 148: Windelband, History of Philosophy, Vol. 1

128 The Greeks : Systematic Period. [Fart L

absolute reality, 1 which, purely for itself, simple and zhangeless,

without origin and imperishable, forms a world by itself, and, as in-

corporeal, is separated from the world where things arise. Hence,

as was demonstrated in the dialogue the Sophist, 2 in a keen polemic

against the doctrine of Ideas, this doctrine formed no principle of

motion, and therefore no explanation of facts, because it excluded

from itself all motion and change.

But however little Plato's interests may have been directed

toward this end, the conception of the Idea as true Being ultimately

demanded, nevertheless, that the phenomenon should be regarded,

not only as something other, something imitative, something that

participated, but also as something dependent. It demanded that

the Idea be regarded as cause of occurrence and change (ama). But

that which is itself absolutely unchangeable and immovable, and

excludes every particular function from itself, cannot be a cause in

the mechanical sense, but only in the sense that it presents the end

for the sake of which the occurrence takes place. Here for the first

time the relation between the two worlds of Being and Becoming

(ova-La and yeVeo-is) is fully defined; all change and occurrence exists

for the sake of the Idea; 3 the Idea is the final cause of phenomena.

This foundation of teleological metaphysics Plato gives in the

Philebus and in the middle books of the Republic, and adds at once a

further culminating thought by introducing as the final cause of all

occurrence, the world of Ideas as a whole, but in particular the high-

est Idea, to which all the rest are subordinate in the sense of means

to end,— the Idea of the Good. This, referring to Anaxagoras, he

designates as the World-reason (yovs), or as the deity}

Side by side with this motif taken from Anaxagoras, another of

a Pythagorean nature appears with increasing force in a later form

of the doctrine of Ideas, a motif in accordance with which the

imperfection of the phenomenon is pointed out as in contrast with

the true Being. This inadequacy, however, could not be derived

from Being itself, and Just as Leucippus, in order to understand

plurality and motion, had declared that in addition to the Being of

1 Symp. 211 B, avrb nad"1 avrb ßed' avrov fwvoeidis del 6v.

2 Page 246 ff. The doctrine there criticised, that of the affdnara et8r], can in

accordance with the individual verbal coincidences be only the Platonic;just

this is a factor in the decision against the genuineness of the dialogue. Schleier-

macher's hypothesis of a Megarian doctrine of Ideas, thought out to rescue the

genuineness, has not shown itself tenable.3 Phileb. 54 C.i Yet we are not to think in this case of personality, or of a spiritual being,

but of the absolute ethical end or purpose of the world, the conception of the.

dyaddv finding an exact definition as little as with Socrates. It is rather presup.

posed as being the simplest, the most comprehensible in itself.

Page 149: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism : Plato. 129

Parmenides the Not-being was also " real," or " actual," and existent,

30 Plato saw himself forced, with like logical consistency, for the

purpose of explaining phenomena and the inadequacy which they

show with reference to the Ideas, to assume beside the world of

Being or of cause, i.e. the world of Ideas and the Idea of the Good,

a secondary or accessory cause (tjvvaiTiov) in that which has not the

attribute of Being. Indeed, the parallelism in the two thinkers

goes so far that this secondary cause, which is not Being (to fxrj 6v),

is for Plato precisely the same as for Leucippus and Philolaus, viz.

empty space.1

Space was then for Plato the " nothing " out of which the world

of phenomena is formed for the sake of the Idea of the Good, or of

the deity. This process of formation, however, consists in taking on

mathematical form; hence Plato taught in the Philebus that the

world of perception was a " mixture " of the "unlimited " (airapov),

i.e. space, and of " limitation " (jripai), i.e. the mathematical forms;

2

and that the cause of this mixture, the highest, divine world-prin-

ciple, was the Idea of the Good. Space assumes mathematical for-

mation in order to become like the world of Ideas.

The importance which mathematics had possessed from the outset

in the development of Plato's thought finds thus at last its metaphys-

ical expression. The mathematical structures are the intermediate

link, by means of which empty space, which is not, is able to imitate

in phenomena the pure "forms" of the world of Ideas. Hence

mathematical knowledge (Siavoia), as well as purely philosophical

knowledge (In-Kmy/ity) , has to do with an abiding essence (ovcrta),

and is therefore comprised together with this, as rational knowledge

(vorjais), and set over against knowledge of phenomena (So£a). But

occupying thus an intermediate place, it takes only the position of a

last stage in the preparation for the wisdom of the " rulers," as set

forth in the system of education in the Republic.

10. The metaphysical preliminaries were now given for what

Plato ultimately projected in the Timceus; viz. a sketch or rough

draught of the philosophy of Nature, for which, of course, true to his

epistemological principle, he could not claim the worth of certainty,

but only that of probability.3 Since, that is, he was not in a position

1 Under the influence of the Aristotelian terminology, this secondary causehas been designated as "matter 1

' (uXij), and it is only recently that modernresearches have made it clear that the Platonic " matter 1

' is simply space. Cf.

H. Siebeck, Untersuchungen z. Philos. d. Gr. (2 Aufl., Freiburg i. B. 1889).2 It is probable that in this case Plato transposed the numbers into the world

of Ideas itself, but looked upon their representation in geometrical structures asthe "limitation" added to space.

3 The Platonic Physics :s then hypothetical in like manner with that of

Parmenides. Here, *oo, it would seem that regard for the demands of his dis*

Page 150: Windelband, History of Philosophy, Vol. 1

130 The Greeks : Systematic Period. [Part I.

to carry through dialectically, and establish in conceptions this

project of explaining occurrence from the world's end or purpose,

Plato gave an exposition of his teleological view of Nature in mythical

form only,— a view intended only as an opinion, and not as science.

This view, nevertheless, takes a position sharply opposed to the

mechanical explanation of Nature, and, as this latter is set forth, wecan scarcely suppose that Plato had any other doctrine in mind than

that of Democritus. In opposition to the theory which makes all

kinds of worlds arise here and there from the "accidental " (mean-

ing "purposeless" or "undesigned") meeting of "that which is in

unordered, lawless motion," and perish again, he sets forth his owntheory that there is only this one, most perfect and most beauti-

ful cosmos, unitary in nature and unique as regards its kind, and

that its origin can be traced only to a reason acting according to

ends.

If, then, it is desired to form a theory concerning this origin, the

ground of the world of phenomena must be sought in the telic rela-

tion of this world to the Ideas. This relation Plato expressed by

the idea of a " world-forming God " (S^iovpyo's, demiurge) whoformed or shaped out that which is not Being, i.e. space, " with

regard to the Ideas." In this connection the Not-being is character-

ised as the indefinite plasticity which takes up all corporeal forms

into itself (Be^a/xevr]) , and yet at the same time forms the ground

for the fact that the Ideas find no pure representation in it. This

counter-working of the accessory cause, or of the individual acces-

sory causes, Plato designates as mechanical necessity (dray*??). Hetakes up then the conception of Democritus as a particular moment

into his physics, in order to explain by it what cannot be under-

stood teleologically. Divine activity according to ends and natural

necessity are set over against each other as explaining principles, on

the one hand for the perfect, and on the other hand for the imper-

fect in the world of phenomena. Ethical dualism passes over from

metaphysics into physical theory.

ciples was united with a polemical purpose. Hence there is found mingled in

the Timceus, a dependence upon Democritus and a combating of his views, anattitude like that of Parmenides toward Heraclitus. Yet the distinction is notto be forgotten, that the Eleatic denied the reality of the world of phenomena,while Plato denied only that it could be known scientifically, i.e. through con-ceptions. In presenting his view, however, Plato goes into questions of astron-

omy, mechanics, chemistry, organic life, physiological psychology, finally eveninto those of medicine. He gives, therefore, a kind of compendious exposi-

tion of his opinions in matters of natural science, opinions which in detail are

extraordinarily fantastic, and as compared with the exact ideas even of his

time, inadequate ; and yet taken in their whole connection, in their relation to

their central principle, they have exercised an effect extending far beyond the

design of their author.

Page 151: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 11.] System of Idealism . Plato. 131

The characteristic fundamental thought of the Platonic as con

trasted with the Atomistic physics is, that while Democritus con-

ceived of the movements of the whole as mechanical resultants of

the original states of motion of the individual atoms, Plato, on the

contrary, regarded the ordered motion of the universe as a whole, as

the primitive unit, and derived every individual change or occur-

rence from this purposively determined whole. From this thought

sprang the strange construction of the conception of the world-soul,

which Plato characterised as the single principle of all motions, and

thus also of all determinations of form, and likewise of all activities

of perception and ideation in the world. 1 In fantastic, obscure ex-

position he brought forward as the mathematical " division " of this

world-soul, his astronomical theory, which was in the main closely

connected with that of the younger Pythagoreans, but which was

less advanced than theirs in its assumption that the earth stood

still. The main criterion in this process of division was the dis-

tinction between that which remains like itself (ravrov) and that

which changes (ddrepov),— a contrast in which we easily recognise

the Pythagorean contrast between the perfect stellar world and the

imperfect terrestrial world.

A similar continuation of Pythagorean doctrine is contained in

the Platonic Timceus, with reference also to the purely mathematical

construction of the corporeal world. Here, too, the four elements

are characterised according to the simple, regular, geometrical solids

(cf. p. 46). But it is expressly taught that these consist of triangu-

lar surfaces, and those, too, of a right-angled sort, which are in part

equilateral, in part so formed that the shorter side is half the length

of the hypothenuse. The limiting surfaces of these solids, — tetrahe-

dron, cube, etc., — maybe thought of as composed of such right-

angled triangles, and Plato would have the essence of space-filling,

i.e. density or solidity of bodies, regarded as consisting in this com-

position of these limiting surfaces. By thus conceiving of physical

bodies as purely mathematical structures, the metaphysical thought

of the Philebus found expression also in physics,— the thought,

namely, that the phenomenal world is a limitation of space formed

in imitation of the Ideas. These triangular surfaces, which were,

moreover, conceived of as being indivisible, have a suspicious simi-

larity with the atomic forms (ax>]fJiaTa ) °f Democritus.

1 In this respect the Timceus, quite as does Democritus, characterises psychical

differences by differences of motion, tracing, for example, right ideation to the

to.vt6i>, merely individual perception to the Odrepov, etc. "Soul" is for the

Greeks at the same time principle of motion and of perception, and just that

(kivt)tlk6i> and aiaOrjTiKbv, Arist. De An. I. 2, 403 b 25), and even Plato makes the

second characteristic dependent upon the first.

Page 152: Windelband, History of Philosophy, Vol. 1

132 The Greeks : Systematic Period. [Part L

§ 12. The Aristotelian Logic.

The breadth of plan which appeared in the systems of the two

great antipodal thinkers, Democritus and Plato, and in accordance

with which their doctrines were methodically developed, made it

indispensable that there should be not only a division of labour, but

a separation of problems. The titles of the writings of Democritus

make it probable that he proceeded clearly and definitely in this

respect also. Plato, to be sure, conceived his literary activity essen-

tially from the artist's point of view, but it is evident that in his

activity as a teacher he did not fail to make that arrangement of

problems for separate treatment which we miss in his dialogues.

In his school the division of philosophy into dialectic, physics,

and ethics became dominant.

If by dialectic in this connection we are to understand essentially

the doctrine of Ideas in its metaphysical development, Aristotle

made the great step in advance of prefacing the investigation of the

subject-matter in all three departments with a preliminary study of

the essential nature of science, a doctrine of the forms and laws of

scientific thought. Even with the Sophists and Socrates reflection

had begun upon the question, in what scientific activity properly

consists, and the sharpened attention given to the inner processes

had made it possible for the abstracting thinker to separate the

general forms of the thought-process itself from the particular con-

tents to which this process relates at different times. All these

beginnings and attempts— for even with Plato it did not go beyond

this— were comprehended by Aristotle in his Logic, and developed

into a complete system in which we have before us the ripe self-

knowledge of Greek science.

1. The immediate aim of the Aristotelian logic is, according to

the express declarations of the philosopher, entirely methodological.

The way is to be shown by which the goal of scientific cognition can

be reached in all departments of knowledge. As in rhetoric the art

of persuasion is taught, so in logic we are to learn the art of scien-

tific investigation, cognition, and proof. For this reason Aristotle

did not reckon logic, which was his greatest creation, among the

philosophical disciplines themselves, but treated it in his lectures

as a propaedeutic, and for this reason his school regarded this

study as the general instrument (opyavov) for all scientific work.

But this preparatory study itself was made a science by Aristotle.

Instead of bringing forward rules of practical value in individual

cases, as may well have been the case with the Sophists, instead

of the general fixing of a principle which had been the service of

Page 153: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 12.] The Aristotelian Logic. 133

Socrates, he offers an examination of the thinking activity on all

sides, a comprehensive examination of its regular forms. He fulfils

the methodological task by formal logic.

But in so doing it becomes evident that the knowledge of the

forms of right thinking can be gained only from understanding the

task of thought, and that in turn this task can be disclosed only

from a definite idea of the general relation of knowledge to its

object. Thus the Aristotelian logic is connected in the most

intimate manner with the metaphysical presupposition which lie

at the basis of his treatment of the other disciplines also. In

its principle, it is thoroughly epistemological.

2. As such, however, it has its roots in the Socratic-Platonic

doctrine of Ideas. That which truly is, is the general or universal,

and knowledge of this is the conception. In this respect Aristotle

always remained a Platonist. What he combated in the system of

his great predecessor 1 was only the Eleatic assumption of absence

of relation,— absence of relation between general and particular,

between Ideas and phenomena, between conceptions and percep-

tions ; an absence of relation which, in spite of all his efforts,

Plato had not overcome, even in the later phase of his teaching.

Even as the final cause of occurrence the Ideas remained a world

by themselves beside (irapa) the phenomena. This tearing apart

(x^pl^eiv) of essence and phenomenon, of Being and Becoming, is?

in addition to special dialectical objections, 2 the object of the chief

reproach which Aristotle brings against the doctrine of ideas.

While Plato had made two different worlds out of the general

which is known by the conception, and the particular which is per-

ceived, the entire effort of Aristotle is directed toward removing

again this division in the conception of reality, and discovering that

relation between Idea and phenomenon which shall make concep-

tional knowledge able to explain what is perceived.

Out of this grows as the primary task for logic, that of recognis-

ing the true relation between the general and the particular, and hence

this fundamental form of abstract or conceptional thought, which

had been already recognised as fundamental by Socrates, stands in

the centre of the Aristotelian logic.

1 Principally in Met. I. 9, and XIII. 4.2 Of these, two are principally worthy of mention in passing. The one

argues, from the logical subordination which obtains among the Ideas, thateverything that we perceive must be subsumed under a number of Ideas ; theother calls attention to the difficulty that the resemblance, which, according tothis system exists between the Idea and the phenomenon, makes necessary still

a higher general above both, etc., in infinitum (ävdpunros— avrdvepuiros— r plrost,vdp(i)TTOs).

Page 154: Windelband, History of Philosophy, Vol. 1

134 The Greeks : Systematic Period. [Part I.

The importance of this same relation grows out of still another

course of thought. If Aristotle found any previous works that

were preparatory for his theory of science, they consisted in the

considerations of the Sophists with regard to the art (principally

rhetorical) of proof and refutation. If now Aristotle asked howone can prove anything scientifically, i.e. in a manner universally

valid and relating to true knowledge, he found that this could con.

sist only in the deduction of the particular from the general. Tq

prove scientifically means to state the grounds for the validity o!f

what is asserted, and these are to be found only in the more general

under which the particular is subsumed.

From this resulted the peculiar complication which constitutes

the Aristotelian conception of science. The general, the Idea, is,

as the true Being, the cause of occurrence and change. It is that,

therefore, out of which and through which the perceived particular

is to be comprehended, conceived, or explained. Science has to set forth

how the perceived particular follows from the general which is

known in conceptions. On the other hand, the general is in thought

the ground by means of which and from which the particular is

proved. Accordingly, conceiving or comprehending and proving are

the same thing, viz. deduction of the particular from the general.

The scientific theory of Aristotle is accordingly concentrated in

the conception of derivation or deduction (cnro'Sei^is). Scientific

explanation of phenomena from true Being is the same logical

process as scientific proof : namely, the deduction or derivation of

what is given in perception from its general ground. Explaining

and proving are therefore denoted by the same word, " deduction,"

and the right proof is that which takes as its ground the actual or

real general cause of that which is to be proved. 1 It is, therefore,

the task of science to exhibit the logical necessity with which the

particular insight (of perception) follows from the general insight (of

conception), and the particular phenomenon from the general cause.

This characterisation of the task of science, thus developed from

metaphysical presuppositions, experienced an essential change in

the progress of its author's investigations.

3. The most immediate task of logic, according to this, is to

establish more exactly what deduction— i.e. on the one hand, proof;

1 This definition of the conception of scientific proof is obviously directedagainst the rhetorical proof of the Sophists. In the art of persuasion, all proofsare welcome, however external they may remain to the true nature of the case,

provided only they are formally sufficient to bring the hearer to assent. Scientific

proof, however, should proceed from the inner, logical necessity of the case, andshould therefore give at the same time insight into the true cause of what is to

be proved.

Page 155: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 12.] The Aristotelian Logic. 135

on the other hand, explanation — properly is, or to set forth those

forms in which thought cognises the dependence of the particular

upon the general. This theory was given by Aristotle in the Analyt-

ics, the logical groundwork, which treats synthetically, in the first

part, of the syllogism, in the second of deduction, proof, and concep-

tion. For in the process of analysing those activities of thought in

which all deduction consists, there results as simple fundamental

form the deduction of one proposition, one statement from another :

i.e. the inference or syllogism (o-vWoyKT/AOs).

The doctrine of the syllogism became thus the central point of the

Aristotelian logic. To this points all that he taught (apparently

only in the most general outlines) concerning the forms of thought

which lie at the basis of the syllogism : out of it come all the points

of view in his methodology.

The outlines of this doctrine, which form the basis of traditional

logic even to this day, are the following. The syllogism is the

deduction of a judgment from two other judgments. Since in a

judgment one concept (the predicate) is affirmed of another concept

(the subject), this affirmation can be grounded only by establishing

the desired connection between the two by means of a third concept,

the middle term (/xe'o-ov). This third concept must then stand in

some relations with the other two, and these relations must be

expressed in two judgments, which are called the premises (77-pora-

<r«s) of the syllogism. Inference, or drawing the conclusion, con-

sists in the process of thought which, from the relations that one

and the same concept (the middle term) sustains to two other

concepts, discovers the relation of these two concepts to each other.

Agreeably to its general presuppositions, the Aristotelian doctrine

of the syllogism fixed its attention upon but one of the possible

relations existing between concepts,— the relation of the subordina-

tion of the particular under the general. The only question for this

theory is always whether the one concept (the subject) should be

subordinated to the other (the predicate) or not. The doctrine of

the syllogism has to do only with the knowledge of those forms

of thought according to which it is to be decided, with the help of

an intermediate concept, whether a subordination of one concept under

another occurs or not. This question Aristotle answered in an abso-

lutely exhaustive manner ; in this consists both the abiding worth

of his doctrine of the syllogism and also the limits of its signifi-

cance.

In correspondence with the fact just noted, Aristotle treats in his

theory of the judgment essentially only the two elements which comeinto consideration for this end : first, Quantity, which determines

Page 156: Windelband, History of Philosophy, Vol. 1

136 The O-reeks : Systematic Period. [Part I.

the kind of subordination of the subject to the predicate as regards

extent, and yields the distinctions of general, particular, and singu-

lar judgments; and second, Quality, according to which this sub-

ordination is either affirmed or denied, and, therefore, the relation

either of connection or of separation is asserted as existing between

the respective extents of the two concepts.

The kinds or figures ( a^rjjxaTa) of the syllogism are, therefore,

essentially fixed by the manner in which the relations of subordina-

tion between the concepts, which are given in the premises, deter-

mine the subordination sought in the conclusion,— a relation which

finds its external expression in the position of the middle term in

the two premises, since this is either the subject of one premise and

predicate of the other, or predicate of both, or subject of both. As

the most valuable and primitive of these three figures, however,

Aristotle consistently designated the first, because in it the principle

of subordination is purely and clearly expressed, since the subject

of the conclusion is subordinated to the middle term, and together

with this, as falling within its compass, is subordinated to the predi-

cate of the major. 1

4. But by defining inference, and so deduction, proof, and expla-

nation in this way, it followed that only propositions of a lesser

degree of generality could be deduced from those of higher generality

by means of this activity so essential to science. That is, by means

of inference, we can never prove anything equally general with the

premises, to say nothing of proving anything more general. Thepeculiar restriction of the ancient idea of the nature of thought,

according to which thought can only apprehend and take apart

what is given but can never produce anything new, makes its

appearance in this feature of the Aristotelian logic. From this,

however, it follows immediately that the deducing, proving, and ex-

plaining science may, indeed, in the individual case, be able to take

that which has served as premise in the syllogism, and deduce it

again as the conclusion of a still more general syllogism, but must,

nevertheless, ultimately proceed from premises which are themselves

capable of no further deduction, proof, and comprehension, of no

reduction to middle terms. The truth of these ultimate premises is,

therefore, immediate (a/xeo-a), not to be deduced, proved or compre-

hended. All deduction needs something primitive ; all proof, a

ground that cannot be proved ; all explaining, something given which

cannot be explained.

1 The details cannot be developed here. Cf. in general, F. Kampe, DieErkenntnisstheorie des Aristoteles (Leips. 1870); R. Eucken, Die Methode der

aristotelischen Forschung (Berlin, 1872).

Page 157: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 12.] The Aristotelian Logic. 137

The apodictic, proving, and explaining activity of science has,

therefore, a limit ; the ultimate grounds of proof are not to be proved;

the ultimate causes used in explaining are not to be explained.

Hence if science is to fulfil its task, which consists in explaining the

particular by means of the general, it must first press forward from

the particular on to the general, in the case of which proving and

explaining are forbidden by the nature of the case, because as imme-

diately certain it asserts itself as not to be deduced and not to be

proved. Hence the processes of deducing, proving, and explaining,

in which the ultimate task of science consists, must be preceded by

the searching out of the starting-points for deduction, of the ultimate

grounds of proof, and of the highest principles of explanation. Theactivity of thought involved in this last process Aristotle calls dia-

lectic, and has laid down its principles in the Topics.

This procedure of searching out the grounds is not, in the nature of

the case, attended by the same " apodictic certainty," as is that of

deducing consequences from the grounds, when the latter are once

established. Investigation proceeds from the particular given in

perception, and from the ideas current in customary opinion (ci/So£ov),

to find the general, from which the particular can then be proved

and explained. Investigatiori, therefore, follows a direction the

reverse of that taken by deduction ; the latter is deductive, the

former inductive, epagogic. The latter proceeds, proving and

explaining, from general to particular ; the former, searching and

testing, from particular to general. 1 Only the completed science is

" apodictic " ; science, in its process of coming into being, is epa-

gogic.

In all these investigations and the contrasts that appear in them,

the chief question for Aristotle is that with regard to judgments

;

but in connection with this he treats also concepts. As a judgment

is proved or deduced, by being concluded from more general judg-

ments, by means of the middle term, so a concept is deduced or

derived by being formed from a more general concept (the next

higher class or genus, yevos) by adding a particular characteristic

mark or difference (&t.a<f>opa). This deduction of the concept is defini-

tion (opta/Aos). As, however, the deduction of propositions ulti-

mately presupposes most general premises, which cannot be further

1 This relation of contrariety between deduction and inquiry Aristotle ex-

pressed in the statements that that which, as regards the nature of the thing, is

the original (irpörepov rfj <pv<rei), and therefore the general, is for human knowl-edge the later, that which must be acquired (vo-repov 7rp6s 17/xas) ; and that, onthe contrary, that which is for us the most immediate (irp6Tepov vpös 17/xäs), the

particular, is, according to the true essence, the derivative, the later (vaTepov t\

iptjcrei).

Page 158: Windelband, History of Philosophy, Vol. 1

138 The Greeks : Systematic Period. [Part L

proved, so, too, definition of lower concepts goes back ultimately to

most general concepts which withdraw from all attempts at deduc-

tion and explanation. These concepts, also, as well as the highest

premises of proof, must be sought inductively; 1 and it seems as

though Aristotle looked upon the propositions of highest generality

as the elucidations of these most general concepts.

5. Among the text-books which Aristotle left, the two main

logical treatises, the Analytics and the Topics, are those which are

most nearly complete by far.2 This may explain the fact that the

logical demands which the Philosopher makes of science are devel-

oped so clearly and surely, while, on the other hand, his system as

carried out in the form known to us, fulfils in but a lesser measure

the expectations thus raised.

For evidently we should expect that a sure statement could be

made as to what the Philosopher declared to be those immediately

certain, highest propositions or concepts which were to be the result

of investigation, and the starting-point of proof and explanation.

If, however, we ask for these, we find ourselves in great embarrass-

ment as regards the teaching of Aristotle. Of general propositions

there is but a single principle, the principle of contradiction? which

he set forth as an unprovable major premise, or highest principle

for all proofs, partly in the purely logical setting that affirmation

and denial of the same combination of concepts reciprocally exclude

each other, partly in the metaphysical form that a thing cannot be

the same and also not be the same. But aside from this he prefers to

call attention to the fact that every department of knowledge has its

own ultimate presuppositions, and does not state these more exactly.

If, however, we seek for the highest concepts,— aside from the

reference made here also to the particular nature of individual dis-

ciplines,— we have the choice between the four "principles" (dpxa0>or " causes," of the Metaphysics, and the " categories," which are

designated as the fundamental forms of predication concerning what

is, — a choice not decided by Aristotle. In both cases we find our-

selves already in the midst of the material as opposed to the formal

elements of his teaching.

1 Over against determination (Trp6a6ecns'), as the deduction of one conceptfrom the higher by adding a new mark, stands therefore abstraction (a<paipe<ris)

as process of formation of class-concepts, — a process which, by continuallytaking away individual characteristics, gains a concept poorer in contents, butwider in its extent. Formation of concepts is, accordingly, with Aristotle, againcompletely analytic, while with Plato it had been intuitive. Aristotle was thefirst to free himself from the optical analogy, in accordance with which the know-ing process of thought had been conceived even by Democritus and Plato.

2 In the case of the Topics, this completeness seems even to have been at-

tained. 3 Met. IV. 3 ff.

Page 159: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 13.] System of -Development : Aristotle. 139

§ 13. The System of Development.

The impression of something completely new, which the logic of

Aristotle makes, as contrasted with all that had previously appeared

in Greek science, rests principally upon the capacity for abstract

thought, presupposed in so high a degree by this separation of the

general Forms of thought from every possible content— a separa-

tion that evinced his genius. This genius for the formation of con-

ceptions by abstraction was evinced by Aristotle in all departments

of his scientific work, and if the " Father of logic " became the

philosophic teacher for two thousand years, he owes this success,

first of all, to the sureness, clearness, and consistency with whichhe formed and defined his conceptions. He fulfilled the task set bySocrates, and in so doing created the language of science. The funda-

mental part of the scientific conceptions and expressions everywhere

in use, even to the present time, goes back to his formulations.

With this inclination to abstraction is connected the further fact

that Aristotle solved the fundamental problem of Greek philosophy

— viz. how behind the changing multiplicity of phenomena a uni-

tary and abiding Being is to be thought— by means of a concept ofrelation, that of development. His two great predecessors had still

been seeking to assign a particular content to the conception of true

Being. Democritus had regarded the atoms and their motion, Plato

the Ideas and their final causation, as the causes of phenomena,—causes different from the phenomena themselves. Aristotle, how-

ever, determined the true reality— that which is

as the essence

zvhich unfolds in the phenomena themselves. He renounced the at-

tempt to think out as the cause of phenomena something different

from them (a second world), and taught that the Being of things

which is known in conception possesses no other reality than the

sum total of the phenomena in which it realises itself. So regarded,

Being (ovala) takes on the character of the essence (to tI rjv emu),

which constitutes the one, only ground of its individual formations,

but is real or actual only in these themselves, and all phenomenal

appearance or coming into being becomes the realisation of the

essence. This is the concept of relation by means of which Aristotle

overcame the opposition of the Heraclitic and Eleatic metaphysics.

1. In particular, the process of development presents itself to

Aristotle as the relation of Form and Matter (efSos, /j-op<f>y— v^n),

Plato ' had declared the world of phenomena to be a mixture of the

1 The main outlines of the Aristotelian metaphysics develop in the simplestway from that phase of the Platonic metaphysics which is presented in the

Philebus (cf. above, § 11, 9). Cf. J. C. Glaser, Die Metaphysik des Aristoteles

(Berlin, 1841).

Page 160: Windelband, History of Philosophy, Vol. 1

140 The Greeks : Systematic Period. [Part I

" unlimited " and of " limitation "; Aristotle holds to the observa-

tion that, in everything of the phenomenal world, formed matter

lies before us. But for him this matter is, indeed, in itself indefi-

nite, and yet not purely indifferent, empty space, but a corporeal

substratum (viroKUfxevov) ; for him, this form is not merely the

mathematical limit, but the form determined as to its contents by

the essence. The matter or material substratum is the possibility

of that which, in the complete thing, has become actual or real by

means of the form. In matter, therefore, the essential nature

(ova-ta) is given only potentially (Swa/xei). First, and only by means

of the form, does it exist in reality or actuality (ivepyeia, actu).

Occurrence, however, or the natural process, is that process in which

the essence passes over from mere possibility, through form, into

actualisation. The essence has not any second, higher reality beside

and apart from the phenomena ; it exists only in the succession of

its phenomenal manifestations, by means of which it realises its

own possibility. The universal is real or actual only in the partic-

ular ; the particular is only because in it the universal realises

itself.

With this transformation of the doctrine of Ideas, Aristotle solves

the fundamental problem of the theoretical philosophy of the

Greeks, viz. that of so thinking Being or what " is " that Becoming,

or the process of Nature (das Geschehen), may be explained from

it. From the Hylozoism of the Milesians on to the opposing

theories of his two great predecessors, all standpoints of Greek

metaphysics are contained as elements in this doctrine of Aristotle.

The Being cognised in conception is the general essence, which

realises itself in its particular phenomenal manifestations from

potentiality on through form, and the process of this realisation is

motion. Being is that which comes to existence in the processes

of Nature. This self-realisation of the essence in the phenomena,

Aristotle calls entelechy (ivreXex^a)

.

2. The central point of the Aristotelian philosophy lies, therefore,

in this new conception of the cosmic processes as the realisation of

the essence in the phenomenon, and the respect in which it is op-

posed to the earlier explanation of Nature consists therefore in

carrying through in conceptions the teleology which Plato had only

set up as postulate, and developed in mythical, figurative form.

While the earlier metaphysics had looked upon the mechanical

process of pressure and impact as the typical fundamental relation

of the cosmic processes, Aristotle regarded as this typical rela-

tion the development of organisms and man's building or forming

activity. From these two departments he took his examples when

Page 161: Windelband, History of Philosophy, Vol. 1

i;hap. ;J, § i-j.j ISystem oj Ueveiopment : Aristotle. 141

he wished to elucidate the metaphysical character of the cosmic

processes. 1

^Nevertheless, the relation of form and matter is not completely

the same in these two kinds of purposive processes, and the differ-

ence between the two asserts itself everywhere in the carrying out

of the Aristotelian fundamental thought. In the case of organic

processes, matter and form are the two sides, separable only through

abstraction, of one and the same reality identical from beginning

to end ; even in the germ which in the process of development

brings the essence to its unfolding, the matter is already shaped

internally by the form. In the case of artistic construction, on the

contrary, the material which contains possibility exists at first by

itself, and the work of the artist with its end in view is added later

to produce the shape by means of motion.

In the latter case, therefore, the development is to be regarded

under four principles. These are the Matter, the Form, the End, and

the Cause of what comes to pass or comes to be.

In the former case, on the contrary, the three other principles, as

set over against the Matter, are but different expressions for the

same thing, since the Form constitutes the Cause and the Result of

the process.

We find, accordingly, that when applied to the task of science,

this fundamental relation of form and matter is carried out in a

twofold way : on the one hand, individual things are regarded as

self-realising forms ; on the other hand, things in relation to one

another are regarded, the one as matter, the other as form. These

two applications of the fundamental principle go through the entire

Aristotelian system side by side, and in the general principles of

the system they sometimes so collide, that it is only by their separa-

tion that apparent contradiction can be cleared away.

3. The former point of view yields the result, that for the Aristo-

telian conception of the world, in contrast with both that of Democ-

ritus and that of Plato, the truly real is the individual thing,

determined in itself by its form. To it, therefore, belongs primarily

the name of essence or substance (ouo-ia). But the essence develops

and realises itself in individual determinations, which are partly its

states (-n-dOrj), partly its relations to other things 2 (ra 717309 n).

Hence knowledge has these which belong to the thing (ra a-vfjLßtßrf-

KoVa) to predicate of it, while the individual thing itself cannot be

predicated of anything else, i.e. in the proposition it can be only

1 Aside from its discussion in the Metaphysics, this question is chiefly treatedIn the Physics.

2 Met. XIV. 2, 1089 b 23.

Page 162: Windelband, History of Philosophy, Vol. 1

142 The G-reeks : Systematic Period. [Part I.

subject and never predicate. 1 Of these modes in which substance

manifests itself, or of the predicates that are possible with regard

to it, Aristotle enumerates as categories, quantity (tto<j6v), quality

(wolov), relation (7rpos n), determination in space and time (77-01), 7n>Te),

action (7rotav), and passion or passivity (Trd<rxeLV ) jand i-n addition,

also, position («eio-ftu) and condition (lx£tl/ )- This collection

(making ten categories inclusive of substance), in which, perhaps,

grammatical observations co-operated, is designed to present the

highest classes or genera under which the contents of all possible

ideas are to be subsumed. Yet Aristotle made no methodical use

of this collection, and his doctrine of the categories acquired, there-

fore, no importance in his metaphysics, aside from the above-noted

relation of substance to its determinations.

When we consider how sharply Aristotle shaped out the scientific

conception of substance in its logical and metaphysical character,

it may appear strange at the first glance that he has announced

neither a methodical principle nor a real principle applying to the

nature of the thing, according to which it would be possible to de-

cide what these truly existing individual things, in his sense of the

word, are. It is clear only that, on the one hand, he did not regard

as essence everything whatever that occasionally appears in ex-

perience as a thing separate from others, and, on the other hand,

that he ascribed this character to organic individuals, to individual

mem It would be in the spirit of his teaching to suppose that he

could have spoken of an " essence " only where an inner determina-

tion of form constitutes the ground of the coherence of individual

characteristics, where, therefore, the knowledge of this essence

solves the problem of science— viz. to determine existent reality

by the general conception — in so far as the abiding individual

thing forms the class-concept for all its particular modes of appear-

ing which show themselves in perception.

But the Socratic-Platonic view of the problem of science brought

•with it the consequence that Aristotle defined yet again the essence

of the individual thing as that through which the individual thing

belongs to its class or species. If substance, as contrasted with its

perceptible phenomena and attributes, presents the universal, on

the other hand the species (ye'vos, or again Platonically, eiSos) is the

universal that realises itself in the individual substances. Here, too,

the same relation is repeated ; the species exists only in so far as it

realises itself in individual things as their truly existing essence,

and the individual thing exists only as the species comes to its phe-

1 Analyt. Post. I. 22, 83 a 24.

Page 163: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 13.] System of Development : Aristotle. 143

nomenal manifestation in it. Just for this reason the species also

have the claim to the metaphysical significance of being essences

(ovaiai). By this means the conception of substance with Aristotle

contains a peculiarly changeable double meaning. The substances

proper are individual things as determined in conception, but as a

second kind of substances (Seirrepcu ova-Lai)l we have the species

which constitute the essence of individual things, just as these latter

constitute the essence of perceptible phenomena.

Scientific knowledge is directed partly toward the conception of

the individual thing, partly toward the conception of the species.

Each of these realises itself in phenomena, and here there is found

much which, as belonging directly to the conception (av/xßeßrjKOTa in

the narrower sense), can be deduced from it, but also much which,

as foreign to the conception, appears in the particular only incident-

ally, as a consequence of the matter in which the conception realises

itself; and of this which is conceptionally indifferent or "accidental "

(o-vfxßeßyjKOTa in the usual sense of the word) there is, according

to the presuppositions of the Aristotelian doctrine, no "theory,"

no scientific knowledge. Hence Aristotle also— and in this lies a

characteristic limit of the ancient study of Nature— disclaimed on

principle any scientific insight into the necessity of law, with which

even the most individual and most particular follow from the gen-

eral. This individual instance he declared rather to be something

really accidental, not to be explained by conception, and limited

scientific consideration to that which is valid universally (i<a$' oXov),

or at least for the most part (e7rl to ttoXv) .

4. In this we see decidedly a holding fast to the tradition of the

doctrine of Ideas : the same attitude discloses itself also in another

direction. If, that is, the relation of matter and form is affirmed

between the different things or classes of things, each of which is

in itself already actual as formed matter, this relation becomes

relative in so far as the same thing which in contrast with a lower

is to be regarded as form, appears as matter when contrasted with

the higher. In this aspect the conception of development becomes

the principle of an ordering of things according to their metaphysical

values, considering these things as rising in uninterrupted succession

from the lowest formations of matter to the highest forms. In

this scale every class of things is assigned its metaphysical dignity

by means of the test that it is regarded as form of the lower and as

the material of the higher.

1 So, at least, they are called in the treatise on categories, the genuineness of

which is, to be sure, not entirely uncontested;yet the designation is quite in

vhe line of Aristotle's teaching taken as a whole.

Page 164: Windelband, History of Philosophy, Vol. 1

144 The Greeks: Systematic Period. [Part J.

This system of individual things, and of their classes, has both a

lower and an upper limit, the former in mere matter, the latter in

pure form. Wholly unformed matter (71-00^77 v\w) is, of course, in

itself, as mere possibility, not actual; it never exists without being

somehow actualised as form. Yet it is not merely that which is not

Being (the Platonic fir/ ov, or empty space), but the accessory cause,

which evinces itself as such through real effects (to ov ovk avev, sine

qua non). Its reality is shown in the fact that the forms do not

completely realise themselves in individual things, and that from it

side-workings (Trapa^vds) proceed which are without connection

with the purposefully active form, or even in contradiction with it.

It is, therefore, from matter that the fact is explained that the

forms realise themselves only potentially (Kara to SuvotöV) : from

matter arises that which is conceptionally indeterminate (o-vfißt-

ßrjKos), or the accidental (avTOfxarov), — the lawless and purposeless

in Nature. Hence the Aristotelian doctrine distinguishes, in its

explanation of Nature, as did Plato in the Philebus, between final

causes (to ov eve«a) and mechanical causes (t6 e£ AvdyKvs) : the former

are the forms which realise themselves in matter ; the latter reside

in matter, out of which proceed side-workings and counter-workings.

Thus the cosmic processes are regarded by Aristotle ultimately

under the analogy of the plastic artist, who finds in the hard material

a limit to the realisation of his formative thought. This material

is, indeed, so far related to the Idea, that the Idea can present itself

in it, at least in general, and yet it is in so far a foreign, and thus

an independent, element, that it in part opposes itself as a retarding

principle to the realising of the forms. Ancient philosophy did not

overstep this dualism between the purposive activity of the form

and the resistance of matter ; with the demand of the teleological

view of the world it united the naive honesty of experience, recog-

nising the necessity, purposeless and contrary to design, which

asserts itself in the phenomena of the actual world.

5. It is, on the contrary, self-evident in the case of pure form,

since its conception is immediately connected with that of true act-

uality, that it possesses in itself the highest actuality without need-

ing any matter whatever. The assumption of such a pure Form is

necessary according to the system of Aristotle, for the reason that

matter, as the merely possible or potential, has in itself alone no

principle of motion or of generation. We cannot, indeed, speak of

a beginning of motion in time in this system of development, which

centres about the conception of self-realising essence, since motion

must be as eternal as Being itself, to the essential characteristics of

which it belongs ; but yet we must point out that property in Being

Page 165: Windelband, History of Philosophy, Vol. 1

Chap. ;5, § 13.] System of Development : Aristotle. 145

which is the cause of motion. This is, however, everywhere the

action of the form upon the matter, in which, with reference to indi-

vidual things, Aristotle distinguishes two elements, viz. an impulse

to be formed inherent in matter, and the purposive motion proceed-

ing from the form itself. But in so far as the form is itself moved,

it must be regarded in turn as matter for a higher form ; and, since

the same thing is true of the latter, and so on, motion would not be

understood if the chain of its causes did not have a first link in the

pure Form which is itself not moved. The first mover (ttpQtov /avow)

is itself unmoved. Hence, in the case of its action upon matter,

only the first of the two elements above mentioned comes into con-

sideration. It operates, not by means of its own activity, but only

by means of the fact that its absolute actuality excites in matter

the impulse to form itself according to it (the prime mover), not as

a mechanical, but as a pure, final cause (kivcI w? ipw/xevov, ob kivov*

fxevov)

.

The prime mover, or the pure Form, means, then, in the Aristo-

telian metaphysics, quite the same thing as the Idea of the Good in

the Platonic, and for it alone Aristotle employs all the predicates

of the Platonic Idea. It is eternal, unchangeable, immovable,

wholly independent, separated (xwpio-rov) from all else, incorporeal,

and yet at the same time the cause of all generation and change.

It is the perfect Being (ivepyeta) in which all possibility is at the

same time actuality ; of all that exists it is the highest (t6 ri fy uvai

to 7rpCirov) and best— the deity. 1

The highest Being or Essence, thus determined according to its

relations, is also characterised by Aristotle as regards its content.

Such an activity, related to no possibility, resting purely within

itself (actus purus), is thought, and thought alone; not, of course,

that mental process which applies itself to individual things and

their changing phenomena, but the pure thought, which is employed

with itself and its eternal nature ; that thought which presupposes

nothing else as an object, but has itself for its constant, unchang-

ing content, the thought of thought (vo'^cns vo^o-ews),— self-conscious-

ness.

In these conceptions, so determined, dwells a significance of

mighty import for the world's history. On the one hand, mono-

1 The exposition of this course of thought from which the later, so-called cos-

mological proof for the existence of God essentially arose, is found principally

in the twelfth book of the Metaphysics. In his popular dialogues Aristotle

amalgamated it with determinations of worth, by giving it the following form :

the distinction between the imperfect and the more perfect which things of

experience show presupposes the reality of a most perfect. Cf. Schol. in Ärist.

487 a 6.

Page 166: Windelband, History of Philosophy, Vol. 1

1 16 The Greeks : Systematic Period. [Part 1.

theism was herewith conceptionally formulated and scientifically

grounded; on the other hand, it passed over from the pantheistic

form, which it had with Xenophanes, and even still with Plato, into

the theistic form, since God is conceived of as a self-conscious being

different from the world. But besides this transcendence, the doc-

trine that God is the absolute mind or spirit (Geist) involves at the

same time the metaphysical advance that the immaterial, the incor-

poreal pure Being, is made equivalent to the spiritual. Spiritual

monotheism is the ripe fruit of Grecian science.

This divine spirituality is conceived of in a purely intellectualistic

manner ; its essential nature is solely thought directed upon itself.

All doing, all willing, is directed toward an object, distinct from the

doer or the wilier. The divine mind, as pure form, needs no object;

he is sufficient for himself, and his knowledge of himself (Oewpia),

which has no other goal than itself, is his eternal blessedness.

He acts upon the world, not through his motion or activity, but

through the longing for him which the world has. The world, and

what takes place in it, arises from the longing of matter after God.

6. Matter (the merely potential) is that which is moved without

itself moving anything; God (the solely actual) is that which moves

without itself being moved; between the two is the entire series of

things, which suffer motion as well as call it forth ; and these, taken

as a whole, are designated by Aristotle as Nature (averts; equivalent

to "world" according to present usage). Nature is, accordingly,

the connected system of living beings viewed as a unity, in which

matter developing ever higher, from form to form, through all the

multitude of its particular shapes, approaches the resting Being of

the deity, and imitating this, potentially takes it up into itself.

But in this connection, the graded scale of things, in the exposition

of which the Aristotelian philosophy of Nature consists, shows a two-

fold standard for estimating relative worth. The scale is therefore

developed in two different series, which find their union only at the

end in a manner which is, indeed, consistent with the fundamental

conceptions of the system, but which is, nevertheless, in itself sur-

prising.

In the conception of the deity, according to Aristotle, there meet,

as chief characteristics, that of Being, resting within itself, and

remaining like itself (MSlov), and that of spirituality or rationality

(voSs). Hence the individual "forms" of Nature take a higher

rank in proportion as they contain the one or the other of these

elements which constitute the highest worth. In the one line,

<:he series of phenomena ascends from the unordered change of the

terrestrial world to the ever-uniform revolution of the stars ; in the

Page 167: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 13.] System of Development : Aristotle. 147

other line, we are led from the merely mechanical change of

place to the activities of the soul and its most valuable develop-

ment, rational knowledge ; and both series have the same terminus,

inasmuch as the stars that are in most uniform motion are con-

ceived of as the highest intelligences, the most rational spirits.

7. In relation to the first of these two aspects Aristotle, taking

up the astronomical views of Plato, adopted the old Pythagorean

antithesis between the earthly and the heavenly world, and it is to

be ascribed to the victorious influence of his philosophy that the

maturer ideas of the later Pythagoreans did not prevail in antiquity,

in spite of their recognition by those learned in astronomy in the

following period. As the whole universe has the most perfect form,

everywhere the same, — that of the sphere,— so among all motions

the most perfect is the circular motion, which returns into itself.

This belongs to the aether, the celestial element, out of which the

stars are formed, and the transparent hollow spheres, in which the

stars move with ever-unchanged uniformity. Farthest out, and in

an absolute changelessness that comes nearest the divine Being, is

the heaven of the fixed stars, beneath that the planets, the sun, and

the moon, whose apparent deviation from the circular movement

was explained by a complicated theory of hollow spheres placed one

within another, the theory which Eudoxus, an astronomer sustaining

a close relation to the Academy, and his disciple Callippus had

propounded. 1 The stars themselves were, however, for Aristotle

beings of superhuman intelligence, incorporate deities. They ap-

peared to him as the purer forms, those more like the deity, and

from them a purposive, rational influence upon the lower life of

earth seemed to proceed, — a thought which became the root of

mediaeval astrology.

The lower "forms" of terrestrial life, on the other hand, are the

four elements (of Empedocles), which are characterised by the ten-

dency to rectilinear motion. But rectilinear motion involves at once

the opposition of two tendencies,— the centrifugal, which belongs to

Fire ; and the centripetal, which belongs to Earth. The first of the

two tendencies is also attributed in a lesser degree to Air, and the

latter in a lesser degree to Water, and so the central mass, our earth,

1 Schiaparelli, Le Sfere Omocentriche di Eudosso, Callippo, ed Aristotele (Mi-lan, 1876). Cf. also O. Gruppe, Die kosmischen Systeme der Griechen (Berlin,

1851). As a principle of method, the following prescription for the proposal of

these questions has been preserved from the Old Academy, typical of the math-ematico-metaphysical presupposition of the speculative explanation of Nature

;

viz. to discover the uniformly ordered motions of the stars by means of whichtheir apparent motions may be explained (dia<rJ}£eiv). Simpl. in Arist. De Coelo

(Karst.), 119.

Page 168: Windelband, History of Philosophy, Vol. 1

148 The Greeks : Systematic Period. [Part I

in a state of rest as a whole, is composed in such a way that about

the earthy material is disposed at first Water and then Air, while

Fire strives toward the celestial outer world. The changing combi-

nations, however, into which the four elements enter, constitute the

imperfect, that which cannot be conceived, that which is accidental

in the terrestrial world. Here the side-working and counter-work-

ing of matter are stronger than in the celestial region where the

mathematical detenu inateness of undisturbed circular motion real-

ises itself.

8. In the changes of the terrestrial world, mechanical, chemical,

and organic processes are built up upon each other in such a waythat the higher always presupposes the lower as its condition.

Without change of place (<t>opd or »«V^o-is in the narrowest sense),

change of qualities (oAaoiWis) is not possible, and the organic

transformation which consists in growth and decay (au^o-is— <f>6i<ns)

is not possible without both the preceding. The higher form is,

however, never merely a product of the lower, but is something self-

subsistent, by means of which those lower forms can be employed

only in a purposive manner.

From this develops an important principle in which Aristotle is

opposed to Democritus, — a principle which the former esteemed

very highly in regard to detailed research in natural science, and

used a great deal, even with express mention. Aristotle l protests

against the attempt to reduce all qualitative to quantitative deter-

minations, — an attempt ultimately accepted even by Plato. Hecombats the contrasting from an epistemological and metaphysical

point of view, of secondary and primary qualities ; to the former he

accords not a less but rather a higher reality than to the latter, and

in the succession of " forms " the inner conceptional character or

determination is evidently of more worth for him than the outer

determination which is capable of mathematical expression. 2 Thf

attempt of Democritus to raise to the rank of a principle for

explaining the world the reduction of all qualitative to quantitative

differences, found its victorious opponent in Aristotle and his doctrine

of the " entelechies," the inner Forms of things. The keen logician

saw that it is never possible to develop qualities analytically from

quantitative relations, and that, on the contrary, the qualit}^ (by which-

ever sense it may be perceived) is something new, which presup-

poses the entire body of quantitative relations as its occasion only.

1 Cf. especially the third book of the treatise De Ccelo.2 For this reason Aristotle also characterises the elements not only by the

different tendencies of their motions, but also by primitive qualities ; and hedevelops them out of a meeting of the contrasted pairs, warm and cold, dry andmoist. Meteor. IV. 1, 378 b 11.

Page 169: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 13.] System of Development : Aristotle. 149

9. With logical consistency the same view is applied by Aristotle

to the relation of the psychical and bodily activities ; the latter are

but the matter for which the former furnish the forms. There is,

with Aristotle, no such dependence of psychical upon corporeal func-

tions as Democritus, in accordance with the procedure of the older

metaphysics, and even Plato, in part (in the Timozus), had taught.

For Aristotle the soul is rather the entelechy of the body, i.e. the

Form which realises itself in the motions and changes of the organic

body. The soul is the cause of bodily formation and motion, a

cause acting from ends ; itself incorporeal, it is yet actual or real

only as the power moving and controlling the body.

But the psychical life itself is also, according to Aristotle, built

up as it were in successive grades or strata, each of which, in turn,

presents matter for the higher. The first Form of organic life is

the vegetative soul {öptirriKov) , which "forms" the mechanical and

chemical changes to the purposive functions of assimilation and

propagation. The soul of plants is restricted to this purely physio-

logical significance of a vital force ; to this is added in the whole

animal kingdom, 1 the animal soul, whose constitutive characteristics

are spontaneous motion in space (kivt/tikoV Kara rbVov) and sensation

(alcrö-qTLKov) .

The purposive, spontaneous motion of the animal body proceeds

from desire (ope£fs), which arises from the feelings of pleasure and

pain, in the form of an effort to procure or shun. But these pre-

suppose everywhere the idea of their object, and are at the same

time bound together with the thought that this object is worthy to

be striven for or to be shunned. The view of the dependence of all

desire upon ideas, peculiar to all Greek psychology, is so strong with

Aristotle, that he even sets forth these relations expressly, accord-

ing to the logical function of judgment, and inference. In the

practical sphere, also, there is affirmation and denial, 2 there is the

process of drawing a conclusion from a general aim to a particular

mode of action.

The proper seat, or home, as it were, of the entire animal life of

ideation is found in sensation. In the physiological psychology

which treats this subject3 Aristotle has used in comprehensive

1 Aristotle's History of Animals (cf. J. B. Meyer, Berlin, 1855) treats in ex-

emplary manner, and with admirable care of detailed investigation, anatomical,

physiological, morphological, and biological problems, and also the questions of

system. The parallel work on plants is indeed lost, but in compensation wehave the work of his friend and disciple Theophrastus.

2 Eth. Nie. VI. 2, 1139 a 21.8 Besides the sections which treat this subject, in the treatise on the Soul, the

smaller treatises attached to this are also to be compared, viz : on Perception,

on Memory, on Dreams, etc.

Page 170: Windelband, History of Philosophy, Vol. 1

150 The Greeks : Systematic Period. [Part I.

manner all the particular information and theories which his prede-

cessors, especially Democritus, possessed on this point; but he

overcame the common inadequacy of all earlier doctrines by conced-

ing a much greater importance to the self-activity of the soul in the

process in which perception arises. Not satisfied to adopt the old

theory that perception consists in a co-operation of object and sub-

ject, he pointed to the unity of consciousness {Einheitlichkeit, fjLto-6rr]<s)

,

with which the animal soul unites what is given in the individual

perceptions of the individual senses to form collective perceptions,

or perceptions that perceive the object as a whole, and in so doing

grasps also the relations of number, situation, and motion. Thus

above the individual senses we must assume the common sense

(koivov alo-OrjTijpLov),1 which is also the seat of recollection, both of

the involuntary or memory (/xi/^/x?/) and the voluntary (ava/iv^orts),

by virtue of the circumstance that in it the perceptions remain as

imaginative representations (<pavTao-Lai) ; at the same time, however,

it is also the seat of our knowledge of our own states. 2

10. Vegetative and animal souls, however, form in man but the

matter for the realisation of the Form peculiar to him,— the reason

(vov<i— Siavoeio-Öcu). By its operation, impulse (ope£t?) becomes will

(ßovX-qo-Ls) ; imaginative representation becomes knowledge (im-

a-Tt]fjir]). It comes as a something new and higher ("from without,"

Bvpadev) to all the psychical activities which develop from perception

even among the beasts. Aristotle expressed this relation by desig-

nating the pure rational activity itself as the active reason (vovs

7roi77TtKos), and, on the contrary, as passive reason (vous TraOrjriKo';),

the material of perceptions, which arises from the bodily existence,

furnishes possibilities and occasions for reason, and is subsequently

worked over and formed by it.

Accordingly the " passive " reason signifies the individual phase

{Erscheinungsweise) given in the natural disposition of the individ-

ual man, and determined by the occasions of his personal experience,

— the " active " reason, on the contrary, signifying the pure reason

considered as a unity in its nature and principles {principielle Ein-

heitlichkeit), common to all individuals. The latter is imperishable,

as it is without beginning, while the former passes away with the

1 With regard to physiological localisation Aristotle found the psychicalactivity to be attached to the vital warmth (e/j.<pvrov 6epn6v), which as animatingbreath {irvevfj.a) is mingled with the blood, and his school developed this doc-trine still further. Cf. H. Siebeck, Zeitschrift für Völkerpsychologie, 1881, pp.36-4 ff. In consequence of this he regarded the heart as the seat of the commonsense and so supplanted the better insight with which Alcmaeon, Diogenes of

Apollonia, Democritus, and Plato had recognised the importance of the brain."2 This beginning for a doctrine of inner perception is found in Arist. De. An.

III. 2, 425 b 12.

Page 171: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 13.] System of Development : Aristotle. 151

individuals in whom it appears. Personal immortality is put in

question by this conclusion just as in the Platonic Timceus, where

it was claimed only for the " rational " " part " of the soul, i.e. that

part which is everywhere alike and impersonal. It is clear that wehave here no longer to do with empirical psychology, but with such

doctrines as have been taken from the systematic connection of

the whole work, and grafted upon psychology in consequence of

ethical and epistemological postulates.

11. In the conception of the reason as the Form peculiar to the

human soul, Aristotle found the key to the solution of that feature

of the ethical problem which even Plato had sought in vain, i.e.

that of the content of the Good. Man's happiness or well-being

(ctiSat/iovta), which in Aristotle's system also is regarded as the

supreme end of all endeavour (re'Xos), is, indeed, dependent in part

upon external fortune ; it is not complete until this has afforded

its good things ; but ethics has to do only with that which stands in

our power (i-a e<£' yfuv), only with the happiness which man gains

by his own activity (irpaKTov ayaOov). Every being, however, be-

comes happy by the unfolding of his own nature and of his ownpeculiar activity — man, therefore, through reason. The virtue of

man is, accordingly, that habitude or permanent state of mind (e£is)

through which he is made capable of the practice of rational activ-

ity ; it develops out of the endowments of his natural disposition,

and has for its fruit, satisfaction, pleasure.

As in the animal soul impulse and perception were to be dis-

tinguished as different expressions, so, too, the reason develops

itself, partly as rational action, partly as rational thought ; as per-

fection, on the one hand, of the character or disposition (^0os), on

the other, of the faculty of intelligence (alo-Odveo-Oai in the broadest

sense of the word). Thus there result, as the excellence or ability

of the rational man, the ethical and the intellectual or dianoetic vir-

tues.

12. The ethical virtues grow out of that training of the will by

which it becomes accustomed to act according to right insight

(<£pov77<ns— opöo? Aoyos). It enables man, in his decisions, to follow

practical reason, i.e. insight into what is correct or proper. Withthis doctrine Aristotle transcends the principles of Socrates, —with evident regard to the facts of the ethical life : not that he

assigned to the will a psychological independence as over against

knowledge ; the point, rather, is, that he gave up the opinion that

the determination of the will arising from rational insight must of

itself be stronger than the desire arising from defective knowledge.

Since experience often shows the reverse of this, man must gain by

Page 172: Windelband, History of Philosophy, Vol. 1

152 The Greeks : Systematic Period. [Part I.

practice that self-control (iyKparaa) by means of which he follows

under all circumstances that which is rationally known, even against

the strongest desires. 1

While to ethical virtue in general belong natural disposition,

insight, and habitude, the individual virtues are distinguished by

the different relations of life to which they refer. A systematic

development of these is not given by Aristotle, but we have, rather,

a comprehensive and delicate treatment of the individual virtues.

The general principle is that rational insight always finds the right

mean between the unreasonable extremes to which the natural

impulsive life leads. Thus courage is the right mean between

cowardice and rashness. A particularly detailed exposition is given

to friendship 2 as the common striving for all that is good and

beautiful, and also to justice as the basis of the political community.

13. For Aristotle, like Plato, was convinced that the moral excel-

lence of man, since it always relates to activities which prosper in

the life of a community, can find its fulfilment only in the life of a

community ; for him, too, there is ultimately no perfect moral life

outside the state, the essential end of which was considered by

Aristotle, also, to be the ethical training of its citizens. As, never-

theless, in the case of the individual man, virtue ought to develop

out of the natural disposition, so the political relations also are

treated by Aristotle from the point of view, that tlie historically

given relations are to be used for the highest possible fulfilment of

that highest end.

Every constitution is right if the government has the ethical weal

of the community as its highest goal ; every constitution has failed

if this is not the case. The good of the state, therefore, does not

depend upon the external form, which is defined by the numberof those who rule. 3 The rule of a single individual may be right

as a kingdom (/WiAeia), bad if a despotism (rupawi's) ; the rule

of few may be good if an aristocracy of culture and disposition,

if an oligarchy of birth or property, bad; the rule of all as a

republic of law and order (7roAtreta) may be good, as mob-rule

(817/xoKpaTta), bad. With profotmd political intelligence, Aristotle

brings together in these expositions the experiences of Grecian

history, and on the ground of the^e enters upon the philosophy of

1 In the polemic against the Socratic doctrine which Aristotle brings forwardin this line, Eth. Nie. III. 1-8, are developed the first beginnings of the problem,of freedom.

2 In the eighth book of the N'icumachosan Ethics.3 A point of view which tbs dialogue the Statesman, passing under Plato's

uame, had already emphasised, while Plato himself in the Republic constructedthe " bad " constitutions fror/j psychological analogies of a predominance of the

iower parts of the soul.

Page 173: Windelband, History of Philosophy, Vol. 1

Chap. 3, § 13.] System of Development : Aristotle. 158

history in giving intimations as to the necessity with which individ-

ual forms of constitutions pass over into one another and. develop

out of one another.

After these presuppositions we can understand that Aristotle

could not think of projecting in detail the constitution of an ideal

state in Plato's manner. He contented himself with a critical

emphasising of those elements which had proved requisite in indi-

vidual constitutions for fulfilling the general task of the state. In

this connection he agrees with the Platonic demand for a public

system of education ; the ethical community must itself take the

care of fitting for their place the elements of which it will in future

consist, and it is the task of education (in the treatment of which

the fragment of the Politics breaks off) to lead man out of his rude

state of nature with the help of the noble arts, to ethical and intel-

lectual culture.

14. To the practical activity of the reason (Aoyio-Ti/cov), in the

broader sense of the word, Aristotle reckoned also "making"(irouh) in addition to "acting" (tt/dS^s)

;yet, on the other hand,

he made so great distinction between this creative activity, whichpresents itself in art, and the action directed toward the ends of

daily life, that he occasionally set the science of art, poietic phi-

losophy, as a third independent science, side by side with the theo-

retical and practical. Of this poietic philosophy, there is preserved

besides the Rhetoric only the fragment of his theory of the art of

poetry, under the name of the Poetic. This sets out, indeed, from

principles relating to the nature of art in general, but in its particu-

lar subject offers only the outlines of a theory of tragedy. In

this, such peculiar relations of this science of art to the two other

principal parts of philosophy appear, that it becomes difficult to sub-

ordinate this branch under either of the other two.

Art is imitative production, and the arts are distinguished as well

by the objects which they imitate as by the material with which

they imitate. The objects of poetic art are men and their actions;

its means are language, rhythm, and harmony. Tragedy, in particu-

lar, represents an important action as performed immediately by

speaking and acting persons. 1

But the purpose of this imitative representation is an ethical one

:

the passions of man, in particular in the case of tragedy, fear and

sympathy, are to be so excited, that by their excitation and en-

hancement purification of the soul (Ka.9a.po is) from these passions

is brought about.

1 Poet. 6, 1449 b 24.

Page 174: Windelband, History of Philosophy, Vol. 1

154 The Greeks : Systematic Period. [Par* I.

On the doctrine of the Catharsis, which became so important for the later

theory of art, and on the literature concerning it, cf. A. Döring, Die Kunstlehredes Aristoteles (Jena, 1876).

The attainment of this end is, however, accomplished in such

a way, that in artistic representation the particular is brought to

our view, not as a particular, but in its universal nature or essence.

Art, like science, has for its object the universal in its particular

realisation ; it offers a kind of knowledge, and with this the pleas-

ure which attends upon knowledge. 1

15. The highest perfection of its development finally is achieved

by the rational nature of man in knowledge. The diano'etic virtues

are the highest, and those which bring complete happiness. Theactivity of the theoretical reason (iTridTrjfxovLKÖv) is directed to the

immediate apprehension of the highest truths, i.e. of the concep-

tions and judgments which the inductive search of scientific inves-

tigation only leads up to without being able to prove, and from

which all deduction must take its beginning (cf. § 12, 4).

But knowledge of these, the full unfolding of the " active reason "

in man, is again designated by Aristotle as a "beholding" (Ocwpla);

and with this beholding of the highest truth man gains a participa-

tion in that pure thought, in which the essence of the deity consists,

and thus, also, in the eternal blessedness of the divine self-conscious-

ness. For this " beholding " which exists only for its own sake

and has no ends of will or deed, this wishless absorption in the

perception of the highest truth, is the blessedest and best of all.

1 Poet. 9, 1451 b 5.

Page 175: Windelband, History of Philosophy, Vol. 1

PART IL

THE HELLENISTIC-ROMAN PHILOSOPHY.

As regards the general literature, the same works serve for this part that were

cited at the beginning of Part I.

With the age of Aristotle, Grecian civilisation stepped out

from its national restrictions and into the great general movementin which the peoples of antiquity that dwelt about the Mediter-

ranean, through interchange and adjustment of their ideas, became

fused into one common civilisation. This process began through

the union of Oriental with Greek thought, in the Hellenistic states

of Alexander's successors. It found its external completion in the

Roman Empire, its internal completion in Christianity. Hellen-

ism, Romanism, and Christianity were the three stages in which the

world's future civilisation developed from antiquity.

The intellectually determining element in this union was Greek

science, and herein consists its significance for the world's history.

It became, like Greek art, the common possession of ancient civili-

sation. To it were joined step by step the highest movements in

the inner life of the peoples, and it became the forming power for

all the longings and impulses that lived within their souls. It

was with the fall of its political independence, with its absorp-

tion into the Empire, that the Greek nation bought the accomplish-

ment of its task of civilisation ; by their dispersal over the world

the Greeks became the teachers of the world.

But in connection with this entrance into more extended relations,

Greek science experienced a separation of the different elements

which were united in it. Together with the purely theoretical

interest in which it had originated, and which had found so clear

an expression in the personality and teaching of Aristotle, a practi-

cal interest had in time developed, which sought in science the

conviction that should govern life. In Plato's philosophy the two

were inseparately fused together, but now these two tendencies of

science became separated.

Scientific thought, which had come to a knowledge of its ownprocesses in the Aristotelian logic, had arrived at the consciousness

155

Page 176: Windelband, History of Philosophy, Vol. 1

156 Hellenistic-Roman Philosophy. [Part II.

of fundamental conceptions, with the aid of which it could use the

abundance of phenomena. The principal opposing theories of the

interpretation of the world had developed in the great systems, and

in this way a fixed frame or setting was formed for the scientific

treatment of detail. But beginning, as it did, with so slightly ex-

tended a knowledge of detail, the more successful Greek science was

in the development of principles, the more it now experienced a

crippling, at once of metaphysical interest and metaphysical force.

In consequence of this, however, the theoretical tendency of sci-

ence was toward details, and the fundamental scientific character of

the Hellenistic-Roman time is erudition and the development of the

specicd sciences. The individual man of science, by entrance into

one of the great schools, gained a firm support of collective opinion,

and a ruling principle for the treatment of separate questions and

subjects which interested him. And indifference toward general

metaphysical theories was the greater, the more it appeared that

fruitful investigation in special provinces, extension of knowledge

of facts, and comprehension of special departments of science were

possible, independently of the strife of metaphysical systems. The

separation of problems, which had been completed typically in the

Aristotelian teaching and school, led necessarily to specialisation,

and the purely theoretical interest in knowledge for its own sake

developed, during the Hellenistic-Roman period, essentially in the

individual sciences. The great savants of later antiquity stand, it is

true, in loose relations with one school or another, but they always

show themselves indifferent to metaphysics. So it happens that

during this time production, so far as the theoretical principles of

philosophy were concerned, was extremely small, while investiga-

tion into mathematics, natural science, grammar, philology, literary

and general history, had rich and comprehensive results to record.

With the great mass of those names which are reckoned as " philos-

ophers," whether heads of schools or associates in the schools, and

which are continued in the schematic treatment of the " History of

Philosophy" only literary-historical notices are connected, as that

they worked specially in this or that department ; or it may be per-

sonal information, of no importance to philosophy, as that they

attached themselves to this or that one among the earlier teachers,

— almost never do we find any formation of new and original con-

ceptions. So far as theoretical knowledge was concerned, this

period turned the old problems of the Greeks hither and thither,

and moved along the track which it found already laid down.

So much the more powerfully, during these centuries of appropri-

ation and elaboration, did the practical significance of philosophy

Page 177: Windelband, History of Philosophy, Vol. 1

Hellenistic-Roman Philosophy. 157

tmfold itself. The need of a scientific doctrine of the ends of

human life, of such a wisdom as should guarantee the happiness of

the individual, could but become more urgent as the ideal structure

of Greek life fell in pieces, as the religion of the people sank ever

more and more to an external tradition, as the crumbling political

life, robbed of its independence, no longer awakened devotion, and

the individual in his inner life felt thrown back upon himself.

Thus wisdom for the conduct of life became the fundamental problem

of the philosophy which followed that of the Greeks, and the nar-

rowing in the statement of the philosophical problem which Socrates,

and after him the Cynic and Cyrenaic schools of Sophistic thought,

had begun, is the general character of the succeeding period.

This did not exclude general theoretical doctrines and their

sharply championed contests from assuming airs of great impor-

tance during this period ; but, on the one hand, they met with no

original interest for their own sake, and consequently developed

only in the directions which were determined by the real end in

view, i.e. that of wisdom for the conduct of life; on the other hand,

they were lacking in originality, they were throughout only the old

traditions shifted about, conditioned by the fundamental practical

thoughts. Even such comprehensive systems as the Stoic and the

Neo-Platonic work only with the conceptions of Greek philosophy,

in order to gain a theoretical basis for their practical ideal. The

key to their theoretical doctrines lies always in the fundamental

practical conviction, and in so far they are all of them character-

istic types of the mingling of problems.

With this predominance of practical importance is connected the

fact that the dependence of philosophy upon the general movement

of civilisation, which had already with the Sophists made its

entrance into the quiet circle of disinterested investigation, became

in the Hellenistic-Roman period a permanent phenomenon, and

this appears most decisively in the changing attitude of this phi-

losophy toward religion.

The development which Greek philosophy had taken, and the

ever more sharply pronounced opposition to the religion of the

people into which it had come, brought with it the result that

the special task of that wisdom for the conduct of life which the

post-Aristotelian philosophy sought, was to find a compensation for

relic/) o us faith. The cultured world, which had lost the support

afforded by religion, and was obliged to give up that of the state

also, sought it in philosophy. As a result, the point of view of the

Hellenistic-Roman wisdom for the conduct of life was primarily

that of individual morality, and the philosophy which busied itself

Page 178: Windelband, History of Philosophy, Vol. 1

158 Hellenistic-Roman Philosophy. [Part II.

with this had, consequently, a thoroughly ethical stamp. The

sharpness of the opposition of this individualistic ethics to religion

appears most clearly among the Epicureans. But in the other

schools, also, the doctrines of the deity have a purely ethical, or

perhaps a theoretical interest, but none that is specifically religious.

This essentially ethical development of philosophy reached its

completion in Greece, especially, indeed, in Athens, which, amid all

the spread of Greek culture eastward and westward, formed for

centuries the centre of scientific life. But soon new centres par-

ticularly for erudite detailed investigation, arose in the great libra-

ries and museums, in Rhodes, in Pergamum, in Alexandria, in

Tarsus, in Rome, and later, in Antioch and Byzantium. Of these,

Alexandria became especially important, where not only did elabora-

tive erudition experience so typical a development, that the entire

direction of this period is generally called " literary-historical " in

accordance with it, but where, also, the philosophical direction of

the time experienced its decided change.

For as time went on philosophy could not remain indifferent to

that deep feeling of dissatisfaction which had seized the ancient

world in the midst of all the glory of the Roman Empire. This

huge empire offered to the peoples which it had welded together

into a mighty unit, no compensation for the loss of their national

independence; it granted them neither inner worth nor outer for-

tune. The draught from the life of earth had become insipid to

ancient peoples, and they thirsted after religion. So they groped

after the different cults and religious practices which individual

peoples had brought with them, and the religions of the Orient

became mixed with those of the Occident.

Into this movement philosophy was the more drawn, the more it

became clear that it could not satisfy the cultured man by the

presentation of its ethical ideal of life,— could not secure for himthe promised happiness. It followed then— at first, in Alexandria

— that the mingling, surging flood of religious ideas emptied itself

into philosophy, which now sought to build up upon a scientific

basis, not only an ethical conviction, but a religion as well. Philos-

ophy employed the conceptions of Greek science to clarify and put

in order religious ideas, to give to the importunate demand of

religious feeling an idea of the world that should be satisfactory

to it, and so created the systems of religious metaphysics, in more or

less intimate connection with the contending religions.

Accordingly, in the Hellenistic-Roman philosophy there are two

distinct periods to be distinguished, the ethical and the religious.

The last century b.c. is to be designated as the time in which the

one gradually passed over into the other.

Page 179: Windelband, History of Philosophy, Vol. 1

CHAPTER I.

THE ETHICAL PERIOD.

The two schools of the great masters of Attic philosophy, the

Academic and the Peripatetic, followed the tendency of the time

which separated science into the two branches, ethical philosophy

and learned investigation. While in the first generation of the

Academy— that contemporary with Aristotle— a Pythagoreanising

metaphysics had predominated, this made room in the next period

for popular moralising (cf. p. 101). In the Lyceum, indeed, Theo-

phrastus, and after him, Strato, held fast to the development and

re-shaping of the Aristotelian metaphysics, but the associates of

Theoprastus, Diccearchus, Aristoxenus, and others, as well as Theo-

phrastus himself, turned to literary-historical studies and to natural

science. Later, the Peripatetics had a great share in the Alexan-

drian erudition, and the history of philosophy especially found in

them its most industrious workers. But in philosophy itself they

played only the conservative role of defending the system of their

school against the attacks of the others, especially upon the ethical

domain, and the new edition of the Aristotelian works by Androni-

cus gave new stimulus for a zealous reproduction of his teaching.

Paraphrases, commentaries, excerpts, and interpretations formed

the chief occupation of the later Peripatetics.

The Academy and Lyceum were, however, injured in their work-

ing by the two schools which were founded toward the end of the

fourth century, and which owed their great success to the fact that

they formulated the tendency of the time toward the practical wis-

dom of life with the clearness and impressiveness of one-sidedness :

namely, the Stoic and the Epicurean.

The first was founded in the Stool ttolklXtj by Zeno, a native of

Oitium in Cyprus, and had, both in his time and in that of his suc-

cessor, Oleanthes, more likeness to Cynicism than in the time of its

third head, Chrysippus, who succeeded in turning the school into a

more scientific course. Epicurus, on the contrary, founded a society

which made the Hedonistic principle, in a refined and intellect-

159

Page 180: Windelband, History of Philosophy, Vol. 1

160 Hellenistic-Roman Philosophy. [Pakt IL

ualised form, its centre, but developed only a slight degree of

scientific vitality. While numerous adherents were won to its

social-ethical principle then established, and to the view of the

world connected with it, as these were continued through antiquity

and especially in the Roman world, the school remained decidedly

more unfruitful scientifically than the others, as well in the special

sciences as in philosophy. Its doctrines have been presented in an

interesting manner by the Roman poet, Lucretius.

These four schools continued side by side in Athens for centuries,

and in the time of the Empire they were still maintained in various

chairs of instruction, and formed there a sort of university ; but

only in the Academy, and here only with great gaps, can a succes-

sion of heads of the school be traced; while the tradition in the

case of the Stoa and the Epicureans breaks off with the first cen-

tury b.c., and for the Lyceum soon after that time.

At first, however, these four schools contended with each other in

the liveliest fashion during the third and second centuries b.c., and

it was especially in ethical questions, and in metaphysical, physical,

and logical questions only in so far as connected with the ethical,

that they sought to bear away the palm from one another. 1

But, moving along side by side with the dogmatic doctrines during

the whole period was another tendency, which, like the Stoic and

Epicurean philosophy, originated in the teaching of the Sophists :

namely, Scepticism. It did not, indeed, take on the form of an

association in a school, but it, too, was brought together into a system-

atic form, and found an ethical culmination. Such a concentration,

in accord with the spirit of the times, of the negative results of the

teaching of the Sophists, was achieved by Pyrrho, whose doctrines

were set forth by Timon. This Sophistical scepticism had the

triumph of obtaining possession of Plato's grove for a time ; for, if

the Middle Academy did not make this doctrine fully its own, it made

it a weapon for combating Stoicism and grounding its own ethics.

In this phase of the development of the Academy appear the two

heads of the school, Arcesilaus and Carneades, who were separated by

about a century. In after time, when the Academy again rejected

Scepticism, this doctrine met with sympathy principally among the

empirical physicians, among whom, even at the end of this period,

u^Jnesidemus and Agrippa are to be mentioned. A complete collec-

tion of the doctrines of the Sceptics, made at a much later time,

is preserved in the works of Sextus Empiricus.

1 Cicero in his philosophical dialogues gives vivid pictures of these school con«

roversies- with a dextrous use of the original sources.

Page 181: Windelband, History of Philosophy, Vol. 1

Chap. l.J The Ethical Period. 161

But the deeper significance of this Scepticism was that it brought

to expression the fundamental frame of mind which had seized the

entire ancient civilisation as it had once seized that of Greece, — a

frame of mind at variance with the true ideal import and content of

that civilisation ; and the same lack of the spirit of decided convic-

tion found only another form in the Eclecticism which began to

develop in the second half of the second century. With the exten-

sion of the schools in the great relations of the life of the RomanEmpire, the school-spirit disappeared, polemic was crippled, and the

need of adjustment and fusion made itself felt instead. The teleo-

logical view of the world, especially, formed the basis upon which

Platonism, Aristotelianism, and Stoicism could agree in a commonopposition against Epicureanism.

The tendency toward such a fusion, toward syncretism, first awokein the Stoic school, and found its most efficient supporters in Pance-

tius and Posidonius, who supplemented the doctrine of the Stoa

on all sides by borrowing Platonic and Aristotelian elements. In

opposition to them stood the New Academy, which, after Philo ofLarissa had made an end of the sceptical episode in the develop-

ment of the school, made the attempt, through Antiochus, to unite

philosophy, then so disunited, upon those doctrines in which Plato

and Aristotle agree.

Less important, because more devoid of principles, but not, there-

fore, the less significant historically, was that sort of eclecticism

which the Romans employed in taking up Greek philosophy. This

consisted in piecing together, from an essentially practical point of

view, the different school systems which met their approval. This

was the case with Cicero, Varro, and in part with the school of the

Sextians.

Of the Peripatetic School (the Lyceum), the co-founder himself is primarily

to be noticed, Theophrastus of Erebus in Lesbos (about 370-287), a somewhatyounger friend of Aristotle, who through his teachings and writings won great

regard for the school. Of his works, the botanical, also a fragment of the

Metaphysics, extracts from his Characters, from the treatise concerning percep-

tion, from his history of physics, and some isolated fragments are preserved(edited by F. Wimmer, Breslau, 1842-62).

With him appear Eudemus of Rhodes, Aristoxenus of Tarentum, whostudied music historically and theoretically {Elemente der Musik, German byR. Westphal, Leips. 1883), Dicaearchus of Messina, a learned polyhistor whowrote a history of Grecian civilisation (ßtos 'EXXdSos), and Strato of Lampsacus,who was head of the school (287-260) and had as surname •' The Physicist."

Among the Peripatetic doxographers, Hermippus, Sotion, Satyrus, HeracleidesLembus (in the second century b.c.), and among the later commentators,Alexander of Aphrodisias (about 200 a.d. in Athens) are to be mentioned.

The Middle Academy begins with Arcesilaus of Pitane in TEolia (about315-241), whose teachings were recorded by his pupil Lacydes, and ends withCarneades (in Rome, 155) and his successor Clitomachus, who died 110. Noth-ing remains of their writings. The sources are, beside Diogeues Laertius, prin-cipally Cicero and Sextus Empiricus.

Page 182: Windelband, History of Philosophy, Vol. 1

162 Hellenistic-Roman Thought : Ethical Period. [Fakt 12.

Just as indirect and general in its character is our knowledge of the NewAcademy. Philo of Larissa was still in Rome in 87. His successor, Antio-chus of Ascalon, was heard by Cicero in Athens in 78. To the supporters of

eclectic Platonism in this first, essentially ethical form belong among othersArius Didymus, who inclined strongly to Stoicism (in the time of Augustus),and Thrasyllus (under Tiberius), who prepared an edition of the works of

Democritus and Plato, arranged according to subjects. An extensive literature

of paraphrase and commentary connected with Plato's works also developed in

the Academy.

When we consider the personality of the Stoic School, we are struck by thefrequency of the descent of its members from the Hellenistic mixed races of theOrient. Thus the founder, Zeno (about 340-265), came from his Cyprian homeas a merchant to Athens, and there, taken captive by philosophy, is said to

have absorbed the doctrines of the different schools, to found his own in the

year 308. His principal pupil was Cleanthes of Assos in Troas, from whosewritings a monotheistic hymn to Zeus is preserved, Stob. Eel. I. 30 (Wachs-muth, p. 25). The scientific head of the school was Chrysippus (280-209) of

Soli or Tarsus in Cilicia. He is said to have written an extraordinary amount,but, aside from the titles, only very unimportant fragments of his works arepreserved. Cf. G. Bagnet (Loewen, 1822). Among the literary-historical

savants of the Stoic School, Diogenes of Babylon and Apollodorus are to bementioned ; Aristarchus and Eratosthenes stood in close relation to the school.

Paneetius (180-110), who was strongly influenced by the Academic scepticism

and who maintained a close relation with the Roman statesmen, began the syn-cretistic development of the Stoa, which was completed by Posidonius of SyrianApamea (about 135-50). The latter was one of the greatest polyhistors of

antiquity, especially in the geographico-historical domain. He taught in Rhodes,and was heard by many young Romans, among whom was Cicero.

Concerning the Stoics of the time of the Empire, cf. the following chapter.

Sources for the Stoic doctrines are Cicero and Diogenes Laertius, Book VII., in

part also the extant writings of the Stoics of the time of the Empire, and thediscoveries at Herculaneum.

IX Tiedmann, System der stoischen Philosophie (3 vols., Leips. 1776) ; P.

Weygoldt, Die Philosophie der Stoa (Leips. 1883) ; P. Ogereau, Essai sur le

Systeme Philosophique des Stoiciens (Paris, 1885) ; L. Stein, Die Psychologieder Stoa (2 vols., Berlin, 1886-88) ;

[Capes, Stoicism, Lond. 1880].

Epicurus (341-270), born in Samos, the son of an Athenian schoolmaster,had already made attempts at teaching in Mitylene and in Lampsacus, beforefounding in Athens, in 306, the society which is named after his "gardens"(ktjttoi, horti, as also the other schools were named after the places where theyassembled). He was much loved as a teacher, on account of his companionablequalities. Of his numerous writings lightly thrown off, the proverbs (/ctfptai

56£ai), three didactic letters, parts of his treatise irepl (pvcreuis (in the discoveries

at Herculaneum), and besides only scattered fragments are preserved ; collected

and arranged systematically by H. Usener, Epicurea (Leips. 1887).Among the great mass of his followers, antiquity brings into prominence his

closest friend Metrodorus of Lampsacus ; also Zeno of Sidon (about 150) andPhsedrus (about 100 b.c.). Philodemus of Gadara in Coele-Syria has become asomewhat more distinct figure to us since a part of his writings has been foundat Herculaneum (Herculanensium voluminum quae supersunt, first series, Naples,

1793 ff.

; second, 1861 ff.) ; the most valuable, irepl arifieluv «cai a-rj/xenia-euiv (cf.

Fr. Bahusch, Lyck, 1879; H. v. Arnim, Philodemea, Halle, 1888).The didactic poem of Tit. Lucretius Carus (98-54), De Natura Berurn, in six

books, has been edited by Lachmann (Berlin, 1850) and Jac. Bernays (Leips.

1852) ;[Eng. ed. with tr. of the poem by Munro, Lond. 1886. Cf. The Atomic

Theory of Lucretius, by J. Masson, Lond. 1884].Further sources are Cicero and Diogenes Laertius, in the tenth book.Cf. M. Guyau, La Morale d'Epicure (Paris, 1878); P. v. Gizycki, Ueber das

Leben und die Moralphilosophie des Epikur (Berlin, 1879) ; W. Wallace, Epi-

cureanism (Lond. 1880) ;[Wallace, Art. Ep. in Enc. Brit. ; W. L. Courtney,

Ep. in Hellenica].

Page 183: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 14.] Ideal of the Sage. 163

Scepticism, as accords with the nature of the case, makes its appearance,

not as a close school, but in looser form.1 It remains doubtful whether the sys-

tematiser of Scepticism, Pyrrho of Elis (perhaps 305-275), had any intimate

relations with the Socratic-Sophistic school of his native city. A certain Bryso,

who passes for the son of Stilpo, is looked upon as an intermediate link. Heaccompanied Alexander on his journey to Asia, together with a follower of

Deinocritus, Anaxarchus by name. The Sinograph, Timon of Phlius (320-230,

the latter part of the time at Athens) from Pyrrho's standpoint derides philoso-

phers. Fragments of his writings in C. Wachsmuth, De Timone Phliasio(Leips. 1859). Cf. Ch. Waddington, Pyrrhon (Paris, 1877).

The external relations of later Scepticism are very obscure and uncertain.

-ffinesidemus from Cnossus taught in Alexandria, and composed a treatise,

Wvppuveioi. \6yoL, of which nothing remains. His life falls probably in the first

century b.c., yet it has also been set almost two centuries later. Of Agrippa,nothing in detail can be established. The literary representative of Scepticism

is the physician Sextus Empiricus, who lived about 200 A.n., and of his writ-

ings there are extant his Outline Sketches of Pyrrhonism (livfipuveioi virorvirüaeis) .

and the investigations comprehended under the name Adversus Mathematicos,of which Books VII.-XI. contain the exposition of the sceptical doctrine, withmany valuable historical notices (ed. by J. Bekker, Berlin, 1842).

Cf. K. Stäudlin, Gesch. und Geist des Skepticismus (Leips. 1794-95) ; N.Maccoll, The Greek Scejytics (London, 1869) ; L. Haas, De PhilosophorumScepticorum Successionibus (Würzburg, 1875) ;

[Owen, Evenings with the Scep-

tics (Lond. 1881) ; A. Seth, Art. Scepticism, in Enc. Brit.'\.

Among the Romans, the admission of philosophy at first encountered violent

resistance ; but by the beginning of the first century b.c. it was the general

custom for the young Romans of superior rank to study in Athens or Rhodes,

and to hear the lectures of the heads of schools, for the same end as that for whichthe Athenians had formerly heard the Sophists. The literary activity of MarcusTullius Cicero (106-43) must be judged from the point of view of his purpose,

which was to awaken among his countrymen an inclination for general scien-

tific culture and a comprehension of its meaning, and from this standpoint his

work is to be highly prized. Skill in composition and grace of form excuse the

lack of proper philosophising ability, which is shown in a selection of doctrines

based on no philosophical principle. The main treatises are De Finibus, DeOfficiis, Tusculance Disputationes, Academica, De Natura Deorum, De Fato,

De Divinatione. Cf. Herbart, lieber die Philosophie des Cicero ; in Works,XII. 167 ff. [Trans, of the above writings of Cicero in the Bohn. Lib.]

His friend, M. Terentius Varro (116-27), the well-known polyhistor andprolific writer, was more learned, but of his labours toward the history of philos-

ophy only occasional notes are extant.

Quintus Sextus and a son of the same name and Sotion of Alexandria are

named as Sextians. Sotion seems to have been the intermediate link in whichthe Stoic morals were brought into union with the Alexandrian Pythagoreanism,

and given that religious turn which characterises them in the time of the Empire.

Some of their Sentences, discovered in a Syrian translation, have been edited

by Gildemeister (Bonn, 1873).On the literary conditions of this whole period cf. R. Hirzel, Untersuchungen

zu Cicero''s philosophischen Schriften (3 vols., Leips. 1877-83).

§ 14. The Ideal of the Wise Man.

The fundamental ethical tendency of the philosophising of this

entire period is still more precisely characterised by the fact that

it is throughout individual ethics that forms the centre of investiga-

tion in this time of epigones. The elevation to the ideals of ethical

1 Hence all reckonings by the successions of heads of the school, attempted

in order to fix the chronology of the later Sceptics, are illusory.

Page 184: Windelband, History of Philosophy, Vol. 1

164 Hellenistic-Roman Thought : JEthical Period. [Part II.

community, in which morals culminated with both Plato and Aris-

totle, was a glorification that had become foreign to its time, of

that through which Greece had become great, viz. the thought

of an active, living state. This had lost power over the hearts of

men, and even in the schools of Plato and Aristotle it found so

little sympathy that the Academicians, as well as the Peripatetics,

brought into the foreground the question of individual happiness

and virtue. What is preserved from the treatise of the Academi-

cian Crantor, On Grief,1 or from the works of Theophrastus under

the title of Ethical Characters, stands wholly upon the footing of a

philosophy that esteems the right appreciation of the good things

of life to be its essential object.

In the endless discussions on these questions in which the schools

engaged in the following centuries, the successors of the two great

thinkers of Attic philosophy found themselves in an attitude of

common opposition to the new schools. Both had pursued through

the entire circuit of empirical reality the realisation of the Idea of

the Good, and in spite of all the idealism with which Plato

especially strove to transcend the world of the senses, they had

not failed to appreciate the relative value of this world's goods.

Highly as they prized virtue, they yet did not exclude the view that

for the complete happiness of man 2 the favour of external fortune,

health, prosperity, etc., are requisite also, and they denied espe-

cially the doctrine of the Cynics and Stoics that virtue is not

only the highest (as they admitted), but also the sole good.

At all events, however, they too laboured to determine the right

conduct of life which promised to make man happy, gnd while

individual members of the schools pursued their special researches,

the public activity, especially that of the heads of the schools

in their polemic with their opponents, was directed to the end of

drawing the picture of the normal man. This it was that the time

desired of philosophy : " Show us how the man must be constituted

who is sure of his happiness, whatever the fortune of the world

may bring him ! " That this normal man must be called the able,

the virtuous, and that he can owe his virtue only to insight, to

knowledge, that he therefore must be the "wise" man,— this is

the presupposition arising from the Socratic doctrine, which is

recognised as self-evident by all parties during this entire period;

and therefore all strive to portray the ideal of the wise man, i.e.

of the man whom his insight makes virtuous, and so, happy.

i Cf. F. Kayser (Heidelberg, 1841).'2 This Aristotelian view was completely assented to by Speusippus and Xen-

ocrates of the Older Academy.

Page 185: Windelband, History of Philosophy, Vol. 1

Chap, l, § 14.] Ideal of the Sage : Epicurus. 165

1. The most prominent characteristic in the conception of the

" wise man," as determined in this period, is, therefore, imperturba-

bility (ataraxy, arapa^ia). Stoics, Epicureans, and Sceptics are un-

wearied in praising this independence of the world as the desirable

quality of the wise man : he is free, a king, a god ; whatever hap-

pens to him, it cannot attack his knowledge, his virtue, his happi-

ness; his wisdom rests in himself, and the world does not trouble

him. This ideal, as thus portrayed, is characteristic of its time;

the normal man, for this period, is not he who works and creates

for the sake of great purposes, but he who knows how to free him-

self from the external world, and find his happiness in himself alone.

The inner isolation of individuals, and indifference toward general

ends, find here sharp expression : the overcoming of the outer world

conditions the happiness of the wise man.

But since he has no power over the world without him, he must

overcome it within himself; he must become master of the effects

which it exercises upon him. These effects, however, consist in

the feelings and desires which the world and life excite in man

;

£hey are disturbances of his own nature— emotions, or passions

(iraOrj, affectus) . Wisdom is shown, therefore, in the relation

which man maintains to his passions. 1 It is essentially freedom

from passions or emotions, emotionlessness (apathy, a-n-a.6i.UL, is the

Stoic expression). To rest unmoved within one's self, this is the

blessing of this " wisdom."

The terms with which this doctrine is introduced in the case of

Epicurus and Pyrrho point immediately to a dependence upon

Aristippus and Democritus. It corresponds to the gradual trans-

formation which took place in the Hedonistic school (cf. § 7, 9) that

Epicurus, 2 who made its principle his own, and likewise designated

pleasure as the highest good, nevertheless preferred the permanent

frame of satisfaction and rest to the enjoyment of the moment.

The Cyrenaics also had found the essence of pleasure in gentle

motion; but— Epicurus held— that is still a "pleasure in motion";

and the state of painless rest, free from all wishes (17801/77 Karao-Tv-

IJL0.TIK7}), is of higher value. Even the zest and spirit of enjoy-

ment has become lost ; the Epicurean would indeed gladby enjoy

1 The ancient conception of the passions {Affect), extending into moderntime (Spinoza), is accordingly wider than that of the present psychology. It is

best denned by the Latin translation u perturbationes animi," "emotions," andincludes all states of feeling and will in which man is dependent upon the outer

world.2 As intermediate links, the younger followers of Democritus, strongly tinc-

tured with Sophistic doctrine, are named; especially a certain Xausiphane-^whom Epicurus heard.

Page 186: Windelband, History of Philosophy, Vol. 1

166 Hellenistic-Roman Thought : JUthical Period. [Part II.

all pleasure, but it must not excite him or set him in motion.

Peace of soul (yaX^noTAo's, cf. § 10, 5) is all that he wishes, and he

anxiously avoids the storms which threaten it, i.e. the passions.

Epicurus therefore recognised the logical consistency with which

the Cynics had characterised absence of icants as virtue and happi-

ness ; but he was far from seriously renouncing pleasure, as they

did. The wise man must, to be sure, understand this also, and act

accordingly, as soon as it becomes requisite in the course of things.

But his satisfaction will be greater in proportion as the compass of

the wishes which he finds satisfied is fuller. Just for this reason,

he needs the insight (<f>p6vr]<n<;) which not only makes it possible to

estimate the different degrees of pleasure and pain as determined

through the feelings, which are to be expected in a particular case,

but also decides whether and how far one should give place to indi-

vidual wishes. In this aspect Epicureanism distinguished three

kinds of wants : some are natural (<£ucra) and unavoidable, so that,

since it is not possible to exist at all without their satisfaction, even

the wise man cannot free himself from them ; others, again, are only

conventional (vo/aw), artificial, and imaginary, and the wise man has

to see through their nothingness and put them from him ; between

the two, however (here Epicurus opposes the radically one-sided

nature of Cynicism), lies the great mass of those wants which have

their natural right, but are not indeed indispensable for existence.

Hence the wise man can in case of necessity renounce them ; but

since the satisfaction of these gives happiness, he will seek to satisfy

them as far as possible. Complete blessedness falls to his lot whorejoices in all these good things in quiet enjoyment, without stormy

striving.

On the same ground, Epicurus prized mental joys higher than

physical enjoyments which are connected with passionate agitation.

But he seeks the joys of the mind, not in pure knowledge, but in

the aesthetic refinement of life, in that intercourse with friends

which is pervaded by wit and sentiment and touched with delicacy,

in the comfortable arrangement of daily living. Thus the wise

man, in quiet, creates for himself the blessedness of self-enjoyment,

independence of the moment, of its demands and its results. Heknows what he can secure for himself, and of this he denies himself

nothing ; but he is not so foolish as to be angry at fate or to lament

that he cannot possess everything. This is his " ataraxy," or im-

passiveness : an enjoyment like that of the Hedonists, but more

refined, more intellectual, and— more blase.

2. Pyrrhd's Hedonism took another direction, inasmuch as he

sought to draw the practical result from the sceptical teachings of

Page 187: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 14.] Ideal of the Sage : Pyrrho, Stoics. 167

the Sophists. According to the exposition of his disciple, Timon, he

held it to be the task of science to investigate the constitution

of things, in order to establish man's appropriate relations to them,

and to know what he may expect to gain from them. 1 But accord-

ing to Pyrrho's theory it has become evident that we can never

know the true constitution of things but at the most can know

only states of feelings (-rrdOr]) into which these put us (Protagoras,

Aristippus). If, however, there is no knowledge of things, it

cannot be determined what the right relation to them is, and

what the success that will result from our action. This scepticism

is the negative reverse side to the Socratic-Platonic inference. Asthere, from the premise that right action is not possible without

knowledge, the demand had been made that knowledge must be

possible, so here the argument is, that because there is no knowl-

edge, right action is also impossible.

Under these circumstances all that remains for the wise man is

to resist as far as possible the seducements to opinion and to action,

to which the mass of men are subject. All action proceeds, as

Socrates had taught, from our ideas of things and their value ; all

foolish and injurious actions result from incorrect opinions. The

wise man, however, who knows that nothing can be affirmed as to

things themselves (d</>acna), and that no opinion may be assented to

(dxaTaX^ia),

2 restrains himself, as far as possible, from judgment,

and thereby also from action. He withdraws into himself, and in

the suspension (i-rroxv)3 oi judgment, which preserves him from

passion and from false action, he finds imperturbability, rest within

himself, ataraxy.

This is the Sceptical virtue, which also aims to free man from the

world, and it finds its limit only in the fact that there are, never-

theless, relations in which even the wise man, withdrawn within

himself, must act, and when nothing else remains for him than to act

according to that which appears to him, and according to tradition.

3. A deeper conception of the process of overcoming the world in

man was formed by the Stoics. At the beginning, to be sure, they

professed quite fully the Cynic indifference toward all goods of the

outer world, and the self-control of the virtuous wise man remained

stamped upon their ethics also as an ineradicable feature ; but they

1 Euseb. Prcep. En. XIV. 18, 2. The doctrine of Pyrrlio is shown by this to

be in exact coincidence with the tendency of the time ;it asks, " What are we

to do, then, if there is no knowledge ?"

2 An expression which was probably formed in the polemic against the Stoic

conception of K<xr6.\t\^n% ; cf. § 17.8 The Sceptics were called also the ifaxriKoL [" Suspenders "] with reference

to this term, characteristic for them.

Page 188: Windelband, History of Philosophy, Vol. 1

168 Hellenistic-Roman Thought : Ethical Period. [Part II.

soon dulled the edge of the radical naturalism of the Cynics by a

penetrating psychology of the impulsive life, which shows a strong

dependence upon Aristotle. They emphasise, still more than the

Stagirite, the unity and independence of the individual soul, as con-

trasted with its particular states and activities, and so, with them,

personality first becomes a determinative principle. The leading-

power, or governing part of the soul (to rjyefxovLKov) , is, for them, not

only that which makes perceptions out of the excitations of the

individual organs in sensation, but also that which by its assent 1

(o-vy/caTaöeo-ts) transforms excitations of the feelings into activities

of the will. This consciousness, whose vocation is to apprehend

and form its contents as a unity, is, according to its proper and

true nature, reason (voCs) ; the states, therefore, in which conscious-

ness allows itself to be hurried along to assent by the violence of

excitement contradict, in like measure, its own nature and reason.

These states (affectus) are, then, those of passion (-n-dör}) and dis

ease of the soul ; they are perturbations of the soul, contrary to

Nature and contrary to reason. 2 Hence the wise man, if he cannot

defend himself from those excitations of feeling in presence of the

world, will deny them his assent with the power of reason ; he does

not allow them to become passions or emotions, his virtue is the

absence of emotions (airäOua.) . His overcoming of the world is his

overcoming of his own impulses. It is not until we give our assent

that we become dependent upon the course of things ; if we with-

hold it, our personality remains immovable, resting upon itself. If

man cannot hinder fate from preparing for him pleasure and pain,

he may, nevertheless, by esteeming the former as not a good, and

the latter as not an evil, keep the proud consciousness of his self-

sufficiency.

Hence, in itself, virtue is for the Stoics the sole good, and on the

other hand, vice, which consists in the control of the reason by the

passions, is the sole evil, and all other things and relations are

regarded as in themselves indifferent (dSidcpopa) .

s But in their

1 This assent, to be sure, even according to the Stoics, rests upon the judg-

ment ; in the case of passion, therefore, upon a false judgment, but it is yet at

the same time the act of the will which is bound up with the judgment. Cf. § 17.2 Diog. Laert. VII. 110: to irddos— i) ä\oyos ko.1 irapä (pvaiv ypvxfy Klvytris rj

6p(j.7) 71-Xeomfouo-a. The psychological theory of the emotions was developedespecially by Chrysippus. Zeno distinguished, as fundamental forms, pleasure

and pain, desire and fear. As principles of division among the later Stoics

there seem to have been used, partly characteristics of the ideas and judgmentswhich call out the emotion, and partly the characteristics of the states of feeling

and will which proceed from it. Cf. Diog. Laert. VII. Ill ff. ; Stob. Eel. II. 174 f.

3 By reckoning even life in this division, they came to their well-knowndefence or commendation of suicide (ifaywyri'). Cf. Diog. Laert. VII. 130;Seneca, Ep. 12, 10.

Page 189: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 14.] Ideal of the Sage : Stoics. 1G9

doctrine of goods they moderate the rigour of this principle by the

distinction of the desirable and that which is to be rejected (nporjy-

fAcva and airo7rpor]yfX€i'a) . Strongly as they emphasised in this con-

nection that the worth (d£ta) which belongs to the desirable is to be

distinguished strictly from the Good of virtue, which is a good in

itself, there yet resulted from this, in opposition to the Cynic one-

sidedness, an at least secondary appreciation of the good things of

life. For since the desirable was valued for the reason that it

seemed adapted to further the Good, and, on the other hand, the

demerit of that which was to be rejected consisted in the hindrances

which it prepares for virtue, the threads between the self-sufficient

individual and the course of the world, which the Cynic paradoxical

theory had cut, were thus more and more knit together again. Themean between what is desirable and what is to be rejected, the abso-

lutely indifferent, survived ultimately only in that which could be

brought in no relation whatever to morality.

As these distinctions, by repression of the Cynic element, gradu-

ally made Stoicism more viable and, so to speak, better able to get

on in the world, so we may see a like modification, by means of

which it became more usable pedagogically, in the later removal of

the abrupt contrast which at the beginning was made between the

virtuous wise and the vicious fools (<f>avXoi, pnnpoC ). The wise man,

so it was said at the beginning, is wise and virtuous entirely, and in

everything the fool is just as entirely and universally foolish and

sinful ; there is no middle ground. If man possesses the force and

soundness of reason, with which he controls his passions, then he

possesses with this one virtue all the individual particular virtues *

at the same time, and this possession, which alone makes happy,

cannot be lost ; if he lacks this, he is a plaything of circumstances

and of his own passions, and this radical disease of his soul commu-

nicates itself to his entire action and passion. According to the

view of the Stoics, therefore, the few sages stood as perfect menover against the great mass of fools and sinners, and in many decla-

mations they lamented the baseness of men with the Pharisaic

pessimism which thus gratifies its self-consciousness. But over

against this first opinion, which looked upon all fools as to be

rejected alike, the consideration presented itself that among these

fools there were always noticeable differences with regard to their

departure from the ideal virtue, and thus between wise men and

fools there was inserted the conception of the man who is progres-

sive and in a state of improvement (-n-poKo-n-Tw). The Stoics, indeed.

1 The Stoics also made the Platonic cardinal virtues the basis for their sys-

'«matic development of their doctrine of the virtues. Stub. Eel. II. 10'J If.

Page 190: Windelband, History of Philosophy, Vol. 1

170 Hellenistic-Roman Thought : ^Ethical Period. [Part II.

held fast to the view that no gradual transition takes place from

this process of improvement to true virtue, and that the entrance

into the condition of perfection results rather from a sudden turn

about. But when the different stages of ethical progress (71-00*0^77)

were investigated and a state was designated as the highest stage,

in which apathy is indeed attained, but not yet with full sureness

and certainty, 1— when this was done, the rigorous boundary lines

were in some measure effaced.

4. Yet in spite of these practical concessions, the withdrawal

of the individual personality within itself remained ultimately an

essential characteristic in the Stoic ideal of life ; on the other hand,

this which these Greek epigones in common regarded as the markof wisdom, was nowhere so valuably supplemented as among the

Stoics. Scepticism, so far as we can see, never desired such a pos-

itive supplementation— consistently enough ; and Epicureanism

sought it in a direction which expressed in the sharpest form the

restriction of ethical interest to individual happiness. For the

positive content of the wise man's peace of soul, hidden from

the storms of the world, is, for Epicurus and his followers, at last

only pleasure. In this they lacked, indeed, that spirited joy of the

sensuous nature with which Aristippus had exalted the enjoyment

of the moment and the joys of the body to be the supreme end,

and we find, as already mentioned, that in their doctrine of the

highest good the blase", critically appreciative epicurism of the culti-

vated man, is declared to be the content of the ethical life. To be

sure, in his psycho-genetic explanation Epicurus reduced all pleasure

without exception to that of the senses, or, as they said later, to

that of the flesh; 2 but, combating the Cyrenaics, he declared 3 that

just these derivative and therefore refined joys of the mind were far

superior to those of the senses. He recognised very properly that

the individual, upon whose independence of the outer world all

hinges, is much surer and much more the master of mental. than

of material enjoyments. The joys of the body depend on health,

riches, and other gifts of fortune, but what is afforded by science

and art, by the intimate friendship of noble men, by the calm, self-

contented and free from wants, of the mind freed from passions,—this is the sure possession of the wise man, almost or wholly un-

touched by the change of fortune. The aesthetic self-enjoyment of

the cultured man is hence the highest good for the Epicureans.

1 Cf. the account (probably with regard to Chrysippus) in Seneca, Ep.75, 8 ff.

2 Athen. XII. 546 (Us. Fr. 409) ; Flut. Ad. Col. 27, 1122 (Us. Fr. 411) ; id.

Contr. Epic. Orat. 4, 1088 (Us. Fr. 429). 3 Diog. Laert. X. 137.

Page 191: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 14.] Ideal of the tiaye : Epicureans, Stoicb. 171

Thus, to be sure, the coarse and sensuous in Hedonism fell away,

and the Gardens of Epicurus were a nursery of fair conduct of life,

finest morals, and noble employments; but the principle of indi-

vidual enjoyment remained the same, and the only difference wasthat the Greeks, in the old age of the national life, together with

their Roman disciples, enjoyed in a more refined, intellectual, and

delicate manner than did their youthful and manly ancestors. Only

the content had become more valuable, because it was the content

presented to enjoyment by a civilisation more richly developed

and deeply lived out; the disposition with which life's cup was

smilingly emptied, no longer in hasty quaffing, but in deliberate

draughts, Avas the same egoism, devoid of all sense of duty. Hencethe inner indifference of the wise man toward ethical tradition and

rules of the land, which we find here also, though with greater cau-

tion ; hence, above all, the putting aside of all metaphysical or

religious ideas that might disturb the wise man in this self-compla-

cent satisfaction of enjoyment, and burden him with the feeling of

responsibility and duty.

5. To this, the Stoic ethics forms the strongest contrast. Already,

in the thought reminding us of Aristotle (§ 13, 11), that the soul

exercises its own proper nature in the rational power with which it

refuses assent to impulses, we may recognise the peculiar antago-

nism which the Stoics assumed in the human psychical life. For

just what we now are likely to call the natural impulses, viz. the

excitations of feeling and will called forth by things of the outer

world through the senses, and referring to these things,— just these

seemed to them, as above mentioned, that which was contrary to

nature (-rrapa. <f>vo-iv). Reason, on the other hand, was for them the

" nature," not only of man, but of the universe in general. When,for this reason, they adopt the Cynic principles in which the moral

is made equivalent to the natural, the same expression contains in

this latter case a completely changed thought. As a part of the

World-reason the soul excludes from itself, as an opposing element,

the determination by sensuous impulses to which the Cynics had

reduced morality : the demands of Nature, identical with those of

reason, are in contradiction with those of the senses.

Accordingly, the positive content of morality among the Stoics

appears as harmony icith Nature, and thus, at the same time, as a

law which claims normative validity as it confronts the sensuous

man (vo/xos).1 In this formula, however, "Nature "is used in a

1 With this is completed an interesting change in Sophistic terminology in

which (§ 7, 1) v6fxos and 6£<ris had been made equivalent to one another, andset over against <pvcns ; with the Stoics v6fj.os = <£ims.

Page 192: Windelband, History of Philosophy, Vol. 1

172 Hellenistic-Roman Thought: Ethical Period. [Part XI.

double sense.1 On the one hand is meant universal Nature, the

creative, cosmic power, the world-thought acting according to

ends (cf. § 15), the Adyos; and agreeably to this meaning, man's

morality is his subordination to the law of Nature, his willing obe-

dience to the course of the world, to the eternal necessity, and in so

far as this World-reason is designated in the Stoic doctrine as deity,

it is also obedience to God and to the divine law, as well as sub-

ordination to the world-purpose and the rule of Providence. The

virtue of the perfect individual, who, as over against other indi-

vidual beings and their action upon him through the senses, ought

to withdraw within himself, his own master, and rest within him-

self, appears thus under obligation to something universal and

all-ruling.

Nevertheless, since according to the Stoic conception the rjye.-

fxoviKov, the life-unity of the human soul, is a consubstantial part of

this divine World-reason, the life in conformity with Nature must

be also that which is adapted to human nature, to the essential

nature of man ; and this, too, as well in the more general sense

that morality coincides with genuine, complete humanity and with

the reasonableness which is valid in like measure for all, as also in

the special meaning, that by fulfilling the command of Nature, each

person brings to its unfolding the inmost germ of his own individual

essence. Uniting these two points of view, it seemed to the Stoics

that a rationally guided consistency in the conduct of life was the

ideal of wisdom, and they found the supreme task of life in this,

that the virtuous man has to preserve this complete harmony with

himself 2 in every change of life, as his true strength of character.

The political doctrinairism of the Greeks found thus its philosophi-

cal formulation and became a welcome conviction for the iron states-

men of republican Rome.

But whatever the particular terms in which the Stoics gave

expression to their fundamental thought, this thought itself waseverywhere the same, — that life according to Nature and according

to reason is a duty (KaOrjKov) which the wise man has to fulfil, %

law to which he has to subject himself in opposition to his sensuous

inclinations. And this feeling of responsibility, this strict conscious-

ness of the " ought," this recognition of a higher order, gives to their

doctrine, as to their life, backbone and marrow.

This demand also, for a life according to duty, we occasionally

meet among the Stoics in the one-sided form, that the ethical con-

1 Cf. Diog. Laert. VII. 87.2 Thus the formulas o/AoXoyovfitvu): rrj <p\jcrei £ijv and o/xoXoyov/xivus $r\v have

ultimately the same meaning. Stob. Eel. II. 132.

Page 193: Windelband, History of Philosophy, Vol. 1

Chaf. 1, § 14.] Ideal of the Sage : Epicureans, Stoics. 173

sciousness requires some things on rational grounds, forbids the

opposites, and declares all else to be ethically indifferent. Whatis not commanded and not forbidden, remains morally indifferent

(äSia<£o/3ov), and from this the Stoics sometimes drew lax conse-

quences, which they perhaps defended more in words than in actual

intention. But here, too, the systematic development of the theory

created valuable intermediate links. For even if only the Good is

unconditionally commanded, yet, in a secondary degree, the desir-

able must be regarded as ethically advisable ; and though baseness

proper consists only in willing that which is unconditionally for-

bidden, the moral man will yet seek to avoid also that which is " to

be rejected." Thus, corresponding to the gradation of goods, there

was introduced a like gradation of duties, which were distinguished

as absolute and " intermediate." So, on the other hand, with regard

to the valuation of human actions, a distinction was made on a somewhat different basis between those actions which fulfil the demandof reason 1 externally— these are called "befitting," conformable to

duty in the broader sense (KaOrjuovTa) — and such as fulfil the de-

mand of reason solely from the intention to do the Good. Only in

the latter case 2is there a perfect fulfilment of duty (KaropOw/xa) ,

the opposite of which is the intention that is contrary to duty, as

evinced in an action, — sin (dfidprrifjia) . Thus the Stoics, proceed-

ing from the consciousness of duty, entered upon a profound and

earnest study, extending sometimes to considerations of casuistry,

of the ethical values of human will and action, and we may regard

as their most valuable contribution the universally applied thought,

that man in all his conduct, outer and inner, is responsible to a

higher command.

6. The great difference in apprehension of the ethical life which

exists between the Epicureans and the Stoics, in spite of a numberof deep and far-reaching common qualities, becomes most clearly

manifest in their respective theories of society and of the state. In

this, to be sure, they are both at one almost to verbal agreement in

the doctrine that the wise man, in the self-sufficiency of his virtue,

needs the state 3 as little as he needs any other society;yes, that in

certain circumstances, he should even avoid these in the interest,

either of his own enjoyment or of the fulfilment of duty. In this

sense, even the Stoics, especially the later Stoics, dissuaded from

1 5aa 6 \6yos alpei voietv; Diog. Laert. VII. 108.2 For the contrast here alluded to by the Stoics Kant has made customary

the expressions legality and morality ; the Latin distinguishes according to

Cicero's precedent, rectum and honestum.8 Epic, in Plut. De Aud. Poet. 14, 37 (Us. Fr. 548).

Page 194: Windelband, History of Philosophy, Vol. 1

174 Hellenistic-Roman Thought : Ethical Period. [Part II.

entrance into the family life and political activity; and for the

Epicureans, the responsibility which marriage and public activity

bring with them was sufficient to justify a very sceptical attitude

toward both, and especially to make the latter appear advisable for

the wise man, only in the case where it is unavoidable, or of quite

certain advantage. In general, the Epicureans hold to the maxim of

their master, to live in quiet, 1 \äßt ßiwo-as, in which the inner crum-

bling of ancient society found its typical expression.

But a greater distinction between the two conceptions of life

shows itself in the fact that, to the Stoics, human society appeared

as a command of reason, which must give way only occasionally to

the wise man's task of personal perfection, while Epicurus expressly

denied all natural society among men,2 and therefore reduced

every form of social conjunction to considerations of utility. So

the theory of friendship, which in his school was so zealously

pledged, even to the point of sentimentality, did not find the ideal

support which it had received in Aristotle's splendid exposition

;

3

it finds ultimately only the motives of the wise man's enjoyment of

culture as heightened in society.4

In particular, however, Epicureanism carried through systemati-

cally the ideas already developed in Sophistic teaching concerning

the origin of the political community from the well-weighed interest

of the individuals who formed it. The state is not a natural structure,

but has been brought about by men as the result of reflection, and

for the sake of the advantages which are expected and received from

it. It grows out of a compact (crvvOrjicrj) which men enter into with

each other in order that they may not injure one another,5 and the

formation of the state is hence one of the mighty processes through

which the human race has brought itself np from the savage state to

that of civilisation, by virtue of its growing intelligence.6 Laws,

therefore, have arisen in every particular case from a convention as

to the common advantage (av^ßokov tov o-u/x^epovros). There is

nothing in itself right or wrong ; and since in the formation of a

compact the greater intelligence asserts itself to its own advantage

1 Plutarch wrote against this the extant treatise (1128 ff.), d Ka\ws Xiyercu

rb \dde /3iw<ras.

2 Arrian, Epic*. Diss. I. 23, 1 (Us. Fr. 525); ib. II. 20, 6 (523).3 Cf. § 13, 12. The extensive literature on friendship is in this respect

a characteristic sign of the time which found its chief interest in the individual

personality and its relations. Cicero's dialogue Lcelius (De Amicitia) repro-

duces essentially the Peripatetic conception.4 Diog. Laert. X. 120 (Us. Fr. 540).5 Cf. among the Kvpiat 86£ai of Epicurus the terse sentences in Diog. Laert. X.

150 f.

6 Cf. the description in Lucretius, De Ber. Nat. V. 922 ff., especially 1103 ff.

Page 195: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 14.] Ideal of the Sage : Stoics. 175

as a matter of course, it is for the most part the advantages of the

wise that disclose themselves as motives in the enaction of laws. 1

And as is the case for their origin and content, so also for their

validity and acknowledgment, the amount of pain which they are

adapted to hinder and pleasure which they are adapted to produce,

is the only standard. All the main outlines of the utilitarian theory

of society are logically developed by Epicurus from the atomistic

assumption that individuals first exist by and for themselves, and

enter voluntarily and with design into the relations of society, only

for the sake of the goods which as individuals they could not obtain

or could not protect.

7. The Stoics, on the contrary, regarded man as already, by virtue

of the consubstantiality of his soul with the World-reason, a being

constituted by Nature for society, 2 and by reason of this very fact

as under obligation by the command of reason to lead a social life,

— an obligation which admits of exception only in special cases.

As the most immediate relation we have here also friendship,

the ethical connection of virtuous individuals who are united in

the common employment of proving in action the moral law.3 But

from these purely personal relations the Stoic doctrine at once passes

over to the most general, to all rational beings taken as an entirety.

As parts of the same one World-reason, gods and men together form

one great rational living structure, a ttoXitlkov avar^ixa, in which

every individual is a necessary member (yuiAos), and from this re-

sults for the human race the ideal task of forming a realm of reason

that shall embrace all its members.

The ideal state of the Stoics as it had been already delineated by

Zeno, partly in a polemic parallel to that of Plato, knows, accord-

ingly, no bounds of nationality or of the historic state; it is a

rational society of all men, — an ideal universal empire. Plutarch,

indeed, recognised 4 that in this thought philosophy constructed as

rational that which was historically prepared by Alexander the

Great, and completed, as we know, by the Romans. But it must

not remain unnoticed that the Stoics thought of this empire only

secondarily as a political power;primarily it was a spiritual unity

of knowledge and will.

It is comprehensible that with such a high-flying idealism the

1 Stob. Flor. 43, 139 (Us. Fr. 530).2 Tuiv (pvaei ttoXltlkQv fäuv : Stob. Eel. II. 226 ff.

3 It was, to be sure, extraordinarily difficult for the Stoics to bring the need,

which they were obliged to recognise as a fact lying at the basis of the social

impulse, into accord with the independence of the wise man, so baldly empha-sised by them.

4 Plut. De Alex. M. Fort. I. 6.

Page 196: Windelband, History of Philosophy, Vol. 1

176 Hellenistic-Roman Thought : Ethical Period. [Part II.

Stoics retained only a very weak interest for actual political life in

the proper sense. Although the wise man was permitted and

indeed charged to take part in the life of some particular state, in

order to fulfil his duty to all even in this base world, yet both the

particular forms of the state and the individual historical states

were held to be ultimately indifferent to him. As to the former,

the Stoa could not become enthusiastic for any of the characteristic

kinds of government, but, following the Aristotelian suggestion, held

rather to a mixed system, something such as Polybius l presented

as desirable on the ground of his philosophico-historical considera-

tion of the necessary transitions of one-sided forms into each other.

To the splitting up of mankind in different states, the Stoics op-

posed the idea of cosmopolitanism, — world-citizenship, — which fol-

lowed directly from their idea of an ethical community of all men.

It corresponded to the great historical movements of the age, that

the difference in worth between Hellenes and Barbarians, which had

been still maintained even by Aristotle,2 was set aside by the Stoics

as overcome,3 and though, in accordance with their ethical principle,

they were too indifferent to the outer relations of position to enter

upon active agitation for social reforms, they demanded, neverthe-

less, that justice and the universal love of man, which resulted as

the highest duties from the idea of the realm of reason, should be

applied also in full measure, even to the lowest members of humansociety— the slaves.

In spite of the fact, therefore, that it turned aside from the

Greek thought of the national state, to the Stoic ethics belongs the

glory that in it the ripest and highest which the ethical life of

antiauitv produced, and by means of which it transcended itself

and. pointed to the future, attained its best formulation. The intrin-

sic worth of moral personality, the overcoming of the world in man's

overcoming of himself, the subordination of the individual to a

divine law of the world, his disposition in an ideal union of spirits

by means of which he is raised far above the bounds of his earthly

life, and yet, in connection with this, the energetic feeling of duty

that teaches him to fill vigorously his place in the actual world,—all these are the characteristics of a view of life which, though

-t'rom a scientific point of view it may appear rather as put together

than as produced from one principle, presents, nevertheless, one of

che most powerful and pregnant creations in the history of the

conceptions of human life.

1 In the extant part of the sixth hook.2 Arist. Pol. 1.2, 1252 b 5.

3 Seneca, Ep. 95, 52 ; cf. Strabo, I. 4, 9. The personal composition also of the

Stoic school was from its beginning decidedly international.

Page 197: Windelband, History of Philosophy, Vol. 1

Chai\ 1, § 14.] Ideal of the Sage : Stoics, Cicero. IT7

8. In a concentrated form all these doctrines appear in the con-

ception of the law of life, determined by Nature and reason for

all men equally, to <f>vaa SUaiov, and this conception, through Cicero, 1

became the formative principle of Roman jurisprudence.

For, in his eclectic attachment to all the great men of Attic phi-

losophy, Cicero not only held fast objectively with all his energy

to the thought of a moral world-order which determines with uni-

versal validity the relation of rational beings to each other, but

he thought also with regard to the subjective aspect of the question

— in correspondence with his epistemological theory (§ 17, 4) —that this command of reason was innate in all men equally, and that

it had grown into inseparable connection with their instinct of self-

preservation. Out of this lex natural, the universally valid natural

law which is exalted above all human caprice, and above all change

of historical life, develop both the commands of morality in general,

and in particular those of human society,— the jus naturale. Butwhile Cicero proceeds to project from this standpoint the ideal form

of political life, the Stoic universal state takes on under his hands 2

the outlines of the Roman Empire. Cosmopolitanism, which had

arisen among the Greeks as a distant ideal, in the downfall of their

own political importance, becomes with the Romans the proud

self-consciousness of their historical mission.

But even in this theoretical development of what the state should

be, Cicero interweaves the investigation of what it is. Not sprung

from the consideration or the voluntary choice of individuals, it is

rather a product of history, and therefore the ever-valid principles

of the law of Nature are mingled in the structures of its life with

the historical institutions of positive law. These latter develop

partly as the domestic law of individual states, jus civile, partly a»

the law which the confederates of different states recognise in their

relation to one another, jus gentium. Both kinds of positive law

coincide to a large extent in their ethical content with the law of

Nature, but they supplement this by the multitude of historical ele-

ments which in them come into force. The conceptions thus formed

are important not only as constructing the skeleton for a new special

science soon to branch off from philosophy ; they have also the

significance that in them the worth of the historical for the first

time reaches full philosophical appreciation : and at this point Cicero

1 Two of his treatises, only partly preserved, come into consideration here,

De TiPpubHrn and De Legibus. Cf. M. Voigt, Die Lphre vom jus naturale, etc.

(Leips. 185C). and K. Hildenbrand, Geschichte und System der Rechts- undStaatsphilosophie, I. 523 ff.

2 Cic. De Eep. II. 1 ff.

Page 198: Windelband, History of Philosophy, Vol. 1

178 Hellenistic-Roman Thought : Ethical Period. [Part II

knew how to transform the political greatness of his people into a

scientific creation.

§ 15. Mechanism and Teleology.

The practice of the schools in the post-Aristotelian period sepa-

rated philosophical investigations into three main divisions,—ethics, physics, and logic (the latter called canonic among the

Epicureans). The chief interest was everywhere given to ethics,

and theoretically the two others were allowed importance only so

far as correct action presupposes a knowledge of things, and this in

turn a clearness with regard to the right methods of knowledge.

Hence the main tendencies of physical and logical theories are

undoubtedly determined in this period by the ethical point of view,

and the practical need is easily contented by taking up and re-shap-

ing the older teachings ; but yet in scientific work the great objects

of interest, especially metaphysical and physical problems, assert

their fascinating power, and so notwithstanding we see these other

branches of philosophy often developing in a way that is not in full

conformity with the nature of the ethical trunk from which they

spring. Particularly in the case of physics, the rich development of

the special sciences must ultimately keep general principles always

alive and in a state of flux.

In this respect we notice first that the Peripatetic School, during

the first generations, made a noteworthy change in the principles for

explaining Nature which it had received from its master.

1. The beginning of this is found already with Theophrastus, whodoubtless defended all the main doctrines of Aristotelianism, espe-

cially against the Stoics, but yet in part went his own ways. The

extant fragment of his metaphysics discusses, among the aporiae,

principally such difficulties as were contained in the Aristotelian

conceptions of the relation of the world to the deity. The Stagi-

rite had conceived of Nature (<£vcris) as a being in itself alive

(£o>ov), and yet had conceived of its entire motion as a (teleological)

effect of the divine Reason ; God, as pure Form, was separated from

the world, transcendent ; and yet, as animating, first-moving power,

he was immanent in it. This chief metaphysical problem of the

following period was seen by Theophrastus, though his own attitude

toward it remained fixed by the bounds of Aristotle's doctrine. Onthe other hand, he shows a more definite tendency in the closely

connected question regarding the relation of reason to the lower

psychical activities. The vovs was regarded, on the one hand (con-

sidered as Form of the animal soul), as immanent, inborn; on the

other hand, in its purity, as different in essence, and as having come

Page 199: Windelband, History of Philosophy, Vol. 1

uhap. 1, § 15. J Mechanism and Teleology : Peripatetics. 179

into the individual soul from without. Here now Theophrastus

decided absolutely against transcendence ; he subsumed the vovs

also as a self-developing activity, under the concept of a cusmic

process, 1 of motion (/aV^o-is), and set it beside the animal soul as

something different, not in kind, but in degree only.

Strato proceeded still more energetically in the same direction.

He removed completely the limits between reason and the lower

activities of ideation. Both, he taught, form an inseparable unity

;

there is no thought without perceptions, and just as little is there

sense-perception without the co-operation of thought ; both together

belong to the unitary consciousness, which he, with the Stoics, calls

to ^yefxoviKov (cf. § 14, 3). But Strato applied the same thought,

which he carried out psychologically, to the analogous metaphysical

relation also. The yjyt^oviKov of the </>uo-is, also, the Reason of Nature,

cannot be regarded as something separated from her. Whether nowthis may be expressed in the form that Strato did not think the

hypothesis of the deity necessary for the explanation of Nature and

its phenomena, or in the form that he postulated Nature itself as

God, but denied it not only external resemblance to man, but even

consciousness, 2— in any case, Stratonism, regarded from the stand-

point of Aristotle's teaching, forms a one-sidedly naturalistic or

pantheistic modification. He denies spiritual monotheism, the con-

ception of the transcendence of God, and by teaching that a pure

Form is as unthinkable as mere matter, he pushes the Platonic

element in the Aristotelian metaphysics, which had remained just

in the thought of the separation (^wpio-^os) of reason from matter,

so far into the background that the element derived from Democ-

ritus becomes again entirely free. Strato sees in what takes place

in the world, only an immanent necessity of Nature, and no longer

the working of a spiritual, extramundane cause.

Yet this naturalism remains still in dependence upon Aristotle, in

so far as it seeks the natural causes of the cosmic processes, not

in the atoms and their quantitative determinations, but expressly

in the original qualities (ttoiott/tcs) and powers (8wa/x«s) of things.

If among these it emphasised especially warmth and cold, this was

quite in the spirit of the dynamic conceptions held by the older

Hylozoism, and to this, also, Strato seems most nearly related in his

undecided, intermediate position between mechanical and teleological

explanation of the world. Just for this reason, however, this side-

development ran its course with Strato himself without further

result, for it was already outrun at the beginning by the Stoic ami

1 Simpl. Fhys. 225 a. 2 Cic. Acad. II. 38. 121 ; De Nat. Deor. I. 18, 35.

Page 200: Windelband, History of Philosophy, Vol. 1

180 Hellenistic-Roman Thought: Ethical Period. [Part II.

the Epicurean physics. These both defended also the standpoint

of the immanent explanation of Nature, but the former was as out-

spokenly teleological as the latter was mechanical.

2. The peculiarly involved position of the Stoics, in the de-

partment of metaphysical and physical questions, resulted from

the union of different elements. In the foreground stands the

ethical need of deducing from a most general metaphysical prin-

ciple the content of individual morality which could no longer find

its roots in state and nationality as in the period of Grecian great-

ness, and therefore of so shaping the conception of this principle as

to make this deduction possible. But, in opposition to this, stood,

as an inheritance from Cynicism, the decided disinclination to regard

this principle as a transcendent, supersensuous, and incorporeal prin-

ciple, out of the world of experience. All the more decisive was the

force with which the thoughts suggested in the Peripatetic philos-

ophy of Nature came forward, in which the attempt was made to

understand the world as a living being, in purposive motion of itself.

For all these motives, the logos doctrine of Heraclitus seemed to

present itself as in like measure a solution of the problem, and this

became, therefore, the central point of the Stoic metaphysics. 1

The fundamental view of the Stoics is, then, that the entire uni-

verse forms a single, unitary, living, connected whole, and that all

particular things are the determinate forms assumed by a divine

primitive power which is in a state of eternal activity. Their doc-

trine is in its fundamental principles pantheism, and (in opposition

to Aristotle) conscious pantheism. The immediate consequence of

it, however, is the energetic effort to overcome the Platonic-Aris-

totelian dualism,2 and remove the opposition between sensuous and

supersensuous, between natural necessity and reason acting accord-

ing to ends, between Matter and Form. The Stoa attempts this

through simple identification of those conceptions whose opposing

characters, to be sure, cannot by this means be put out of the world.

Hence it declares the divine World-being to be the primitive

power in which are contained in like measure the conditioning laws

and the purposeful determination of all things and of all cosmic

processes, — the World-ground and the World-mind. As actively

productive and formative power, the deity is the Aoyos a-n-epfxaTiKo?,

1 Cf. H. Siebeck, Die Umbildung der peripatetischen Naturphilosophie in die

der Stoiker {Unters, z. Philosophie der Griechen, 2 Aufl., pp. 181 ff.).

2 If we were obliged to conceive of the relation of Aristotle to Plato in asimilar manner (§ 13, 1-4), just in this point the Stoic philosophy of Natureshows a farther development in the same direction which the Peripatetic takesin Strato.

Page 201: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 15.] Mechanism and Teleology : Stoics. 181

the vital principle, which unfolds itself in the multitude of phenom-

ena as their peculiar, particular Ao'yoi cnrepfjLaTiKoi or formative forces.

In this organic function, God is, however, also the purposefully

creating and guiding Reason, and thus with regard to all particular

processes the all-ruling Providence (717)01/010.). The determination of

the particular by the universe (which constitutes the dominant

fundamental conviction of the Stoics) is a completely purposeful

and rational order,1 and forms as such the highest norm (vo/xos),

according to which all individual beings should direct themselves in

the development of their activity. 2

But this all-determining " law " is for the tStoics, as it was for

Heraclitus, likewise the all-compelling power which, as inviolable

necessity (avayKrj), and so, as inevitable destiny (elp.app.evv, fatum),

brings forth every particular phenomenon in the unalterable succes-

sion of causes and effects. Xothing takes place in the world with-

out a preceding cause (cuVia Trpo-qyovfxivw), and just by virtue of this

complete causal determination of every particular does the universe

possess its character of a purposeful, connected whole.3 Hence

Chrysippus combated in the most emphatic manner the conception

of chance, and taught that apparent causelessness in a particular

event could mean only a kind of causation hidden from humaninsight. 4 In this assumption of a natural necessity, admitting of no

exceptions even for the most particular and the least important

occurrence,— a conviction which naturally found expression also in

the form that the divine providence extends even to the smallest

events of life,5— the Stoic school agrees even verbally with Democ-

ritus, and is the only school in antiquity which carried this most

valuable thought of the great Abderite through all branches of

theoretical science.

In all other respects, indeed, the Stoics stand in opposition to

Democritus and in closer relation to Aristotle. For while in the

Atomistic system the natural necessity of all that comes to pass

results from the motive impulses of individual things, with the

Stoics it flows immediately from the living activity of the icliole, and

1 As the Platonic Timfeus had already taught, §11, 10.2 The normative character in the conception of the logos appeared clearly even

with Heraclitus (§ 6, 2, p. 03, note 5).3 Plut. De Fa to, 11, 574. * lb. 7, 572.8 Plutarch makes Chrysippus say (Comm. Not. 34, 5, 1070) that not even the

meanest thing can sustain any other relation than that which accords with the

decree of Zeus. Cf. Cic. De Nat. Deor. II. 05, 1(54. Only the circumstancethat the Stoa limited the immediate action of the divine providence to the pur-

poseful determination of the whole, and derived from this that of the particular,

explains such modes of expression as the well-known Magna dii curant, parvanegligunt. Cf. § 10. 3-

Page 202: Windelband, History of Philosophy, Vol. 1

182 Hellenistic-Roman Thought : Ethical Period. [Part IX

as over against the reduction of all qualities to quantitative differ-

ences, they held fast to the reality of properties as the peculiar

forces of individual things, and to qualitative alteration (dAAoiWis,

in opposition to motion in space). They directed their polemic

particularly against the purely mechanical explanation of natural

processes by pressure and impact ; but in carrying out their teleology,

they sank from the great conception of Aristotle, who had every-

where emphasised the immanent purposiveness of the formations in

which the Forms were realised, to the consideration of the benefits

which flow from the phenomena of Nature to meet the needs of

beings endowed with reason, " of gods and men." l In particular,

they exaggerated, even to ridiculous Philistinism, the demonstration

of the manner in which heaven and earth and all that in them is,

are arranged with such magnificent adaptation for man}3. In all these theoretical views, and just in these, the Epicureans

are diametrically opposed to the Stoics, With the Epicureans, em-

ployment with metaphysical and physical problems had in general

only the negative purpose 3 of setting aside the religions ideas

through which the quiet self-enjoyment of the wise man might be

disturbed. Hence it was the chief concern of Epicurus to exclude

from the explanation of Nature every element that would allow a

government of the world, guided by universal ends, to appear as

even possible ; hence, on the other hand, the Epicurean view of the

world was absolutely lacking in a positive principle. This explains

the fact that Epicurus, at least, had only a sceptical shrug of the

shoulders for all questions of natural science from which no practical

advantage was to be gained ; and though many of his later disciples

seem to have been less limited, and to have thought more scien-

tifically, the ruts of the school's opinion were worn too deep to

allow the attainment of essentially broader aims. The more the

teleologieal conception of Nature formed, in the course of time, the

common ground on which Academic, Peripatetic, and Stoic doctrines

met in syncretistic blending, the more Epicureanism insisted upon

its isolated standpoint of negation ; theoretically, it was essentially

anti-teleological, and in this respect brought forth nothing positive.

It was successful only in combating the anthropological excres-

cences to which the teleologieal view of the world led, especially

1 Cic. De Fin. III. 20, 67 ; De Nat. Deor. II. 53 ff.

2 If one might trust Xenophon's Memorabilia, the Stoics had in this no less

a man than Socrates as their predecessor; yet it seems that even in this account,

which is tinctured with Cynicism if not worked over from the Stoic point of

view (Krohn), the general faith of Socrates in a purposeful guiding of the worldby divine providence has descended into the petty. Cf. § 8, 8.

3 Diog. Laert. X. 143 ; Us. p. 74.

Page 203: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 15.] Mechanism and Teleology : Epicureans. 188

with the Stoics, 1— a task which was undoubtedly not so very diffi-

cult,— but to create from principles a counter-theory it was not pre-

pared. Epicurus, indeed, availed himself for this purpose of the

external data of the materialistic metaphysics, as he was able to

receive them from Democritus ; but he was far from attaining the

latter's scientific height. He could follow the great Atomist cnly

so far as to believe that he himself also, for explaining the world,

needed nothing more than empty space and the corporeal particles

moving within it, countless in number, infinitely varied in form and

size, and indivisible ; and to their motion, impact, and pressure he

traced all cosmic processes, and all things and systems of things

(worlds) which arise and again perish, thereby seeking to deduce

all qualitative differences from these purely quantitative relations. 2

He accepted, accordingly, the purely mechanical conception of nat-

ural processes, but denied expressly their unconditioned and excep-

tionless necessity. The doctrine of Democritus, therefore, passed

over to the Epicureans only in so far as it was Atomism and mechan-

ism ; with regard to the much deeper and more valuable principle

of the universal reign of law in Nature, his legacy, as we have seen

above, passed to the Stoics.

Meanwhile, just this peculiar relation is most intimately con-

nected with the Epicurean ethics and with the decisive influence

which that exercised upon their physics ; indeed, one may say that

the individualising tendency taken by the ethical reflection of the

post-Aristotelian age found its most adequate metaphysics just in

the doctrine of Epicurus. To a morals, which had for its essential

content the independence of the individual and his withdrawal

upon himself, a view of the world must have been welcome which

regarded the prime constituents of reality as completely independ-

ent, both of each other and of a single force, and regarded their

activity as determined solely by themselves.3 Now the doctrine of

Democritus which taught the inevitable, natural necessity of all

that comes to pass, contains unmistakably a (Heraclitic) element

which removes this autonomy of individual things, and just to their

adoption of this element did the Stoics owe the fact (cf. § 14, 5)

that their ethics outgrew the one-sided Cynic presuppositions with

which they started. It is all the more comprehensible that Epi-

curus let just this element fall away; and his conception of the

1 Cf. especially Lucret. De Her. Nat. T. 1021 ; V. 156 ; Diog. Laert. X. 07.2 Sext. Emp. .4*'. Math. X. 42.3 Thus Epicurus grounded his deviation from Democritus's explanation of the

world by an appeal to human freedom of the will. Cf. § 10, and also the cita*

tions in Zeller IV. 3 408, 1 [Eng. tr. Stoics, etc., p. 446].

Page 204: Windelband, History of Philosophy, Vol. 1

184 Hellenistic-Roman Thought . Ethical Period. [Part II.

world as contrasted with that of the Stoa is characterised precisely

by this, that while the latter regarded every individual as deter-

mined by the whole, he rather regarded the whole as a product of

originally existing and likewise originally functioning individual

things. His doctrine is in every respect consistent Atomism.

Thus the system of Democritus had the misfortune to be propa-

gated for traditions of antiquity, and so also for those of the Middle

Ages, in a system which indeed retained his Atomistic view, looking

in the direction of the exclusive reality of quantitative relations

and of the mechanical conception of the cosmic processes, but

set aside his thought of Nature as a connected whole, regulated by

law.

4. Following this latter direction, Epicurus gave a new form to

the doctrine of the origin of the world maintained by Atomism. 1

In contrast with what had been already seen, perhaps by the

Pythagoreans, but, at all events, by Democritus, Plato, and Aristotle,

that in space in itself there is no other direction than that from

the centre toward the periphery, and the reverse, he appeals to the

declaration of the senses, 2— agreeably to his doctrine of knowl-

edge, — according to which there is an absolute up and down, and

maintains that the atoms were all originally in motion from above

downward by virtue of their weight. But, in order to derive the

origination of atom groups from this universal rain of atoms, he

assumed that some of them had voluntarily deviated from the direct

line of fall. From this deviation were explained the impacts, the

grouping of atoms, and, ultimately, the whirling motions which

lead to the formation of worlds, and which the old Atomism had

derived from the meeting of atoms which were moving about in an

unordered manner.3

It is noteworthy, however, that after he had in this way spoiled

the inner coherence of the doctrine of Democritus, Epicurus re-

nounced the voluntary choice of the atoms as a means for the

further explanation of the individual processes of Nature, and from

the point when the whirling motion of the atom-complexes seemed

to him to be explained, allowed only the principle of mechanical

1 Ps.-Plut. Plac. I. 3 ; Box. D. 285 ; Cic. Be Fin. I. 6, 17 ; Guyau, Moraled'Epic. 74.

2 Diog. Laert. X. 60.3 Cf. § 4, 9. It seems that later Epicureans who held fast to the sensuous

basis of this idea and yet would exclude the voluntary action of the atoms andcarry out more thoroughly the Democritic thought of Nature's conformity to

law, hit upon the plan of explaining the grouping (äßpoLo-/j.6s) of the atoms onthe hypothesis that the more massive fell faster in empty space than the •' lighter";

at least, Lucretius combats such theories (Be Her. Nat. II. 225 ff.V

Page 205: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 15.] Mechanism and Teleology : Epicureans. 185

necessity to stand. 1 He used, therefore, the voluntary self-determi-

nation of the atoms only as a principle to explain the beginning of

a whirling motion which afterwards went on purely mechanically.

He used it, therefore, just as Anaxagoras used his force-matter, vovs

(cf. p. 52). For upon this metaphysical substructure Epicurus

erected a physical theory which acknowledged only the mechanicsof atoms as explanation for all phenomena of Nature without anyexception, and carried this out, for organisms especially, by employ-

ing for the explanation of their purposive formation the Empedo-clean thought of the survival of the fit.

Lastly, the Democritic principle of natural necessity asserts itself

in the system of Epicurus in his assumption that in the continuous

arising and perishing of the worlds which become formed by the

assemblages of atoms, every possible combination, and thus every

form of world-construction, must ultimately repeat itself. This

was proved in a manner which would now be put upon the basis of

the theory of probabilities, and the result of this repetition washeld to be, that considering the infinitude of time, nothing can

happen which has not already existed in the same way. 2 In this

doctrine, again, Epicurus agrees with the Stoics, who taught a plu-

rality of worlds, not co-existent, but following one another in time,

and yet found themselves forced to maintain that these must be

always completely alike, even to the last detail of particular forma-

tion and particular events. As the world proceeds forth from the

divine primitive fire, so it is each time taken back again into the

same after a predetermined period : and then when after the world-

conflagration the primitive power begins the construction of a newworld, this «^w« (Nature), which remains eternally the same, unfolds

itself again and again in the same manner, in correspondence with

its own rationality and necessity. This return of all things (TraAiy-

yevecria or d7ro/caTao-Tacns) appears, accordingly, as a necessary con-

sequence of the two alternative conceptions of the Stoics, Aoyos and

el/xapfi€vrj-

5. The theoretical ideas of these two main schools of later an-

tiquity are accordingly at one only in being completely material-

1 Hence in a certain sense it might be said, from the standpoint of presentcriticism, that the difference between Democritus and Epicurus was only a rela-tive one. The former regards as an unexplained primitive fact the directionwhich each atom has from the beginning, the latter regards as an unexplainedprimitive fact a voluntary deviation, taking place at some point of time, from adirection of fall which is uniform for all. The essential difference, however, is

that with Democritus this primitive fact is something timeless, while withEpicurus it is a single voluntary act occurring in time, an act which is expresslycompared with the causeless self-determination of the human will (cf. § IG).

2 Plut. in Euseb. Dox. D. 581, 19 ; Us. Fr. 266.

Page 206: Windelband, History of Philosophy, Vol. 1

186 Hellenistic-Roman Thought : Ethical Period. [Part II.

istic, and it was just in opposition to Plato and Aristotle that

they expressly emphasised this position of theirs. Both maintain

that the real (to. ovto), because it manifests itself in action and

passion {ttoluv Kai 7racr^etv) , can be only corporeal ; the Epicureans

declared only empty space to be incorporeal. On the contrary,

they combated the (Platonic) view that the properties of bodies

are something incorporeal per se (kciÖ' ea-uro),1 and the Stoics even

went so far as to declare that even the qualities, forces, and rela-

tions of things, which present themselves in changing modes in

connection with things and yet as actual or real, are " bodies," 2

and with a mode of thought which reminds us of the coming and

going of the homoiomerise with Anaxagoras, 3 they regarded the

presence and change of properties in things as a kind of inter-

mixture of these bodies with others, a view from which resulted

the theory of the universal mingling and reciprocal interpenetration

of all bodies («pao-is oV oXwv).

In carrying out the materialistic theory the Epicureans produced

scarcely anything new ; on the contrary, the Stoic doctrine of Nature

shows a number of new views, which are interesting not only in

themselves, but also as having marked out the essential lines for

the idea of the world held during the following centuries.

First of all, in the Stoic system the two antitheses, which were

to be removed or identified in the conception of Nature as one, again

part company. The divine primitive essence divides into the active

and the passive, into force and matter. As force, the deity is fire

or warm, vital breath, pneuma ; as matter, it changes itself out of

moist vapour (air) partly into water, partly into earth. Thus fire is

the soul, and the " moist " is the body, of the World-god ; and yet

the two form a single being, identical within itself. While the

Stoics thus attach themselves, in their doctrine of the transmuta-

tion and re-transmutation of substances, to Heraclitus, and in their

characterisation of the four elements principally to Aristotle, and

follow Aristotle also in the main in their exposition of the world-

structure and of the purposive system of its movements, the most

important thing in their physics is doubtless the doctrine of the

pneuma.

God as creative reason (Aoyos o-jrep/xaTiKos) is this warm vital

breath, the formative fire-mind which penetrates all things and is

1 Diog. Laert. X. 67.

2 Plut. C. Not. 50, 1085.3 A similar materialising of the Platonic doctrine of Ideas (Plat. Phwdo, 102),

which reminds ns of Anaxagoras, was apparently worked out by Eudoxos, whobelonged to the Academy (p. 103). Arist Met. I. 9, 991 a 17, and also Alex.

Aphr. Schol. in Arist. 573 a 12.

Page 207: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 15.] Mechanism and Teleology : Epicureans, Stoics. 187

dominant in them as their active principle ; he is the universe

regarded as an animate being, spontaneously in motion within

itself, and purposefully and regularly developed. All this is

comprehended by the Stoics in the conception of the -n-vcvixa,1 an

extraordinarily condensed conception, full of relations, — an idea

in which suggestions from Heraclitus (Xoyos), Anaxagoras (voik),

Diogenes of Apollonia (arjp), Democritus (fire-atoms), and not least

the Peripatetic natural philosophy and physiology, became intri-

cately combined. 2

6. The most effective element in this combination proved to be

the analogy between macrocosm and microcosm, universe and man,

which the Stoics adopted from Aristotle. The individual soul, also,

the vital force of the body, which holds together and rules the flesh,

is fiery breath, pneuma; but all the individual forces which are

active in the members and control their purposive functions, are

also such vital minds or spirits (spiritus animates) . In the humanand the animal organism the activity of the pneuma appears con-

nected with the blood and its circulation ; nevertheless, the pneuma

itself— just because it is also a body, said Chrysippus 3— is sep-

arable in detail from the lower elements which it animates, and this

separation takes place in death.

At the same time, however, the individual soul, as it is only a

part of the universal World-soul, is completely determined in its

nature and its activity by this World-soul ; it is consubstantial with

the divine Pneuma and dependent upon it. Just for this reason the

World-reason, the Xoyos, is for the soul the highest law (cf. above,

§ 14, 3). The soul's independence is therefore only one that is

limited by time, and in any case it is its ultimate destiny to be

taken back into the divine All-mind at the universal conflagration

of the world. With regard to the continuance of this independence,

i.e. as to the extent of individual immortality, various views were

current in the school; some recognised the duration of all souls

until the time of the universal conflagration, others reserved this

for the wise only.

As now the one Pneuma of the universe (whose seat was located

by the Stoics sometimes in heaven, sometimes in the sun, sometimes

in the midst of the world) pours itself forth into all things as

animating force, so the ruling part of the individual soul (to yjye^o-

vlkov or Xoyio-fxv;) in which dwell ideas, judgments, and impulses, and

1 Stob. Eel. I. 374. Box. D. 463, 16: elvai rb ov irveviia. klvovv eavrb irpbs

iavrb Kal e£ avrou, 77 wvevfj-a eavrb klvovv irpbaco Kai dirtcrw kt\.2 Cf. H. Siebeck Zeitsch. f. Völkerpsychologie, 1881, pp. 364 ff.

8 Nemesius, Be Nat. Horn. p. 34.

Page 208: Windelband, History of Philosophy, Vol. 1

18S Hellenistic-Roman Thought : HJthical Period. [Fart II.

as whose seat the heart was assumed, was regarded as extending its

particular ramifications throughout the whole body, like the " arms

of a polyp." Of such particular "pneumata" the Stoa assumed

seven,— the five senses, the faculty of speech, and the reproductive

power. As the unity of the divine Primitive Being dwells in the

universe, so the individual personality lives in the body.

It is characteristic that the Epicureans could entirely adopt

this external apparatus of psychological views. For them, too, the

soul— which according to Democritus consists of the finest atoms—is a fiery, atmospheric breath (they apply likewise the term

"pneuma"); but they see in this breath something that is intro-

duced into the body from without, something held fast by the body

and mechanically connected with it, which in death is forthwith

scattered. They also distinguish between the rational and the

irrational part of the soul, without, however, being able to attribute

to the former the metaphysical dignity which it acquired in the Stoic

theory. Here, too, their doctrine is, on the whole, insufficient and

dependent.

7. In accordance with the pantheistic presupposition of the

system, the metaphysics and physics of the Stoics form also a

theology, a system of natural religion based on scientific demonstra-

tion, and this found also poetic presentations in the school, such as

the hymn of Cleanthes. Epicureanism, on the contrary, is in its

whole nature anti-religious. /It takes throughout the standpoint of

" Enlightenment," that religion has been overcome by science, and

that it is the task and triumph of wisdom to put aside the phantoms

of superstition which have grown out of fear and ignorance.^/ Thepoet of this school depicts in grotesque outlines the evils' which

religion brought on man, and sings the glory of their conquest byscientific knowledge. 1 It is all the more amusing that the Epicurean

theory itself fell to depicting a mythology of its own which it re-

garded as harmless. It believed that a certain degree of truth must

attach to the universal faith in gods, 2 but it found that this correct

idea was disfigured by false assumptions. These it sought in the

myths which feigned a participation of the gods in human life, and

an interference on their part in the course of things ; even the

Stoics' belief in Providence appeared to them in this respect as but

a refined illusion. Epicurus, therefore, — following Democritus in

his doctrine of the eidola, or images (§ 10, 4),— saw in the gods

giant forms resembling men, who lead a blessed life of contemplation

and spiritual intercourse in the intermediate spaces between the

i Lucret. De Ber. Nat. I. 62 ff. 2 T)iog. Laert. X. 123 f . ; Us. p. 59 f.

Page 209: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 1.3.J Mechanism and Teleoloyy : Epicureans, tftoics. 189

worlds (intermundia), undisturbed by the change of events, and

unconcerned as to the destiny of lower beings ; and thus this doc-

trine, also, is fundamentally only the attempt of Epicureanism to

put in mythological form its ideal of aesthetic self-enjoyment.

8. It was in an entirely different way that the ideas of the

popular religion were fitted into the Stoic metaphysics. Whereas,

up to this time in the development of Greek thought philosoph-

ical theology had separated itself farther and farther from the

indigenous mythology, we meet here, for the first time, the

systematic attempt to bring natural and positive religion into

harmony. Accordingly, when the Stoics, also, yielded to the need of

recognising the warrant of ideas universally present throughout the

human race (cf. § 17, 4), their pneuma doctrine offered them not

only a welcome instrument, but suggestions that were determinative.

For consideration of the universe must teach them that the divine

World-power has evidently taken on mightier forms and those of

more vigorous life than individual human souls ; and so, beside the

one deity without beginning and end, which for the most part they

designated as Zeus, a great number of "gods that had come into exist-

ence," made their appeaxanee. To these the Stoics, as Plato and

Aristotle had already done, reckoned first of all the stars, which

they too honoured as higher intelligences and especially pure for-

mations of the primitive fire, and further, the personifications of

other natural forces in which the power of Providence, benevolent

to man, reveals itself. From this point of view we can understand

how an extensive interpretation of myths was the order of the day

in the Stoic school, seeking to incorporate the popular figures in its

metaphysical system by all kinds of allegories. In addition to this

there was an equally welcome use of the Euemeristic theory, which

not only explained and justified the deification of prominent men,

but taught also to consider the demons sacred, as the guardian

spirits of individual men.

Thus the Stoic world became peopled with a whole host of higher

and lower gods, but they all appeared as ultimately but emanations of

the one highest World-power, — as the subordinate powers or forces

which, themselves determined by the universal Pneuma, were con-

ceived of as the ruling spirits of the world's life. They formed,

therefore, for the faith of the Stoics, the mediating organs, which

represent, each in its realm, the vital force and Providence of the

World-reason, and to them the piety of the Stoics turned in the

forms of worship of positive religion. The polytheism of the popular

faith was thus philosophically re-established, and taken up as au

integrant constituent into metaphysical pantheism.

Page 210: Windelband, History of Philosophy, Vol. 1

190 Hellenistic-Roman Thought : ^Ethical Period. [Part II.

In connection with this scientific reconstruction of positive rev

ligion stands the theoretical justification of divination in the Stoic

system where it awakened great interest, except in the case of a few

men like Panaetius, who thought more coolly. The interconnection

and providentially governed unity of the world's processes was held to

show itself— as one form of manifestation— in the possibility that

different things and processes which stand in no direct causal rela-

tion to one another, may yet point to one another by delicate rela-

tions, and therefore be able to serve as signs for one another. Thehuman soul is capable of understanding these by virtue of its rela-

tionship with the all-ruling Pneuma, but for the full interpretation

of such ecstatic revelations the art and science of divination, resting

upon experience, must be added. On this basis Stoicism regarded

itself as strong enough to elaborate philosophically all the divination

of the ancient world. This was especially true of its younger repre-

sentatives, and in particular, as it seems, of Posidonius.

§ 16. The Freedom of the Will and the Perfection of the World.

The sharp definition of the contrasted mechanical and teleological

views of the world, and especially the difference in the conceptional

forms in which the thought, common to a certain extent, of Nature's

universal conformity to law had been developed, led, in connection

with the ethical postulates and presuppositions which controlled

the thought of the time, to two new problems, which from the

beginning had various complications. These were the problems of

the freedom of the human will and of the goodness and perfection

of the world. Both problems grew out of contradictions which

made their appearance between moral needs and just those meta-

physical theories which had been formed to satisfy those needs.

1. The proper home for the formation of these new problems

was the Stoic system, and they may be understood as the necessary

consequence of a deep and ultimately irreconcilable antagonism be-

tween the fundamental principles of the system. These principles are

metaphysical monism and ethical dualism. The fundamental moral

doctrine of the Stoics, according to which man should overcome the

world in his own impulses by virtue, presupposes an anthropological

duality, an opposition in human nature in accordance with which

reason stands over against a sensuous nature contrary to reason.

Without this antithesis the whole Stoic ethics is ready to fall. The

metaphysical doctrine, however, by which the command of reason

in man is to be explained, postulates such an unrestricted and all-

Page 211: Windelband, History of Philosophy, Vol. 1

Chap. 1, § lti.j Freedom of the Will : Socrates, Aristotle. 191

controlling reality of the World-reason that the reality of what is

contrary to reason, either in man or in the course of the world,

cannot be united therewith. From this source grew the two ques-

tions which since then have never ceased to employ man's critical

investigation, although all essential points of view that can comeinto consideration in the case were more or less clearly illumined at

that time.

2. The conceptions which form the presuppositions for the prob-

lem of freedom lie ready at hand in the ethical reflections on the

voluntary nature of wrongdoing, which were begun by Socrates

and brought to a preliminary conclusion by Aristotle in a brilliant

investigation. 1 The motives of these thoughts are ethical through-

out, and the domain in which they move is exclusively psychologi-

cal. The question at issue is hence essentially that of freedom of

choice, and while the reality of this is doubtless affirmed upon the

basis of immediate feeling, and with reference to man's conscious-

ness of his responsibility, difficulty arises only in consequence of

the intellectualistic conception of Socrates, who brought the will

into complete dependence upon insight. This difficulty develops

primarily in the double meaning of " freedom," or, as it is here still

called, " voluntariness " (Ikovo-iov), an ambiguity which has since

been repeated again and again in the most variously shifted forms.

According to Socrates, all ethically wrong action proceeds from a

wrong view— a view clouded by desires. He who thus acts does

not " know," therefore, what he is doing, and in this sense he acts

involuntarily. 2 That is, only the wise man is free ; the wicked is

not free.3 From this ethical conception of freedom, however, the

psychological conception of freedom— i.e. the conception of freedom

of choice as the ability to decide between different motives— must

be carefully separated. Whether Socrates did this is a question ;

*

at all events, it was done by Plato. The latter expressly affirmed

man's freedom of choice, 5 appealing to his responsibility, — a psycho-

logical decision on essentially ethical grounds,— and, at the same

time, he held fast to the Socratic doctrine that the wicked man acts

involuntarily, i.e. is ethically not free. He even connects the two

directly when he develops the thought 6 that man may sink into the

1 Eth. Nie. III. 1-8.2 Xen. Mem. III. 9, 4 ; Cyrop. III. 1, 38.3 Cf. Arist. Eth. Nie. III. 7, 113 b 14.4 According to a remark in the Peripatetic Magna Moralia (I. 9, 1187 a 7)

Socrates, indeed, had expressly said, " it is not in our power" to be good or badAccording to this, therefore, he had denied psychological freedom.

5 Vl&t.Itep. X. 617 ff.

6 Plat. Fhced. 81 B.

Page 212: Windelband, History of Philosophy, Vol. 1

2y2 Hellenistic-Roman Thought : Ethical Period. [1-art IL

condition of ethical non-freedom by his own fault, and, therefore,

with psychological freedom.

With Aristotle, who separated himself farther from the Socratic

intellectualism, the psychological conception of freedom comes out

more clearly and independently. He proceeds from the position

that ethical qualification in general is applicable only in the case of

"voluntary " actions, and discusses in the first place the prejudices

which this voluntariness sustains, partly from external force {ßia)

and psychical compulsion, and partly from ignorance of the matter.

That action only is completely voluntary which has its origin in the

personality itself, and of which the relations are fully known. 1 The

whole investigation 2is maintained from the standpoint of responsi-

bility, and the discovered conception of voluntariness is designed to

lead to the conception of accountability. It contains within itself

the characteristics of external freedom of action, and of a conception

of the situation unclouded by any deception. But, on this account,

it must be still further restricted, for among his voluntary acts a

man can be held accountable for those only that proceed from a

choice (Trpoat'/aeo-ts).3 Freedom of choice, therefore, which proceeds

by reflecting upon ends as well as upon means, is the condition of

ethical accountability.

Aristotle avoided a farther entrance upon the psychology of

motivation and upon the determining causes of this choice ; he con-

tents himself with establishing the position that the personality

itself is the sufficient reason for the actions 4 which are ascribed to

it ; and to this maintenance of the freedom of choice his school, and

especially Theophrastus, who composed a treatise of his own on

freedom, held fast.

3. On this same basis we find also the Stoics, in so far as purely

ethical considerations are concerned. Precisely that lively feeling

of responsibility which characterises their morals demanded of them

the recognition of this free choice on the part of the individual, and

they sought therefore to maintain this in every way.

Their position became critical, however, by reason of the fact

that their metaphysics, with its doctrine of fate and providence,

drove them beyond this attitude. For since this theory of fatf

made man, like all other creatures, determined in all his external

and internal formation and in all that he does and suffers, by the

1 Eth. Nie. III. 3, 1111 a 73 : 01/17 apxh &v o-vt£ el56ri to. ko.6' 'iKaara ev oh i)

.'/>a£ts.

2 As the reference at the beginning to the right of punishment clearly shows[Eth. Nie. 1109 b34).

8 lb. 4, 1112 a 1.

4 lb. 5, 1112 b 31 : ioiKe Sr] . . . ävdpuTos eivai äpxi T& v irpd^euv.

Page 213: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 16.] Freedom of the Will : Stoics, Epicurus. i9<?

aii-aniinating World-power, personality ceased to be the true ground

(apxv) °f his actions, and these appeared to be, like all else that

occurs, but the predetermined and unavoidably necessary operations

of the God-Nature. In fact, the Stoa did not shrink from this

extreme consequence of determinism ; on the contrary, Chrysippus

heaped up proof on proof for this doctrine. He based it upon the

principle of sufficient reason (cf. above, § 15, 2) ; he showed that

only by presupposing this could the correctness of judgments con-

cerning the future be maintained, since a criterion for their truth or

falsity is given only if the matter is already determined ; * he also

gave to this argument the changed form, that since only the

necessary can be known, and not that which is still undecided, the

foreknowledge of the gods makes necessary the assumption of deter-

minism ; he even did not scorn to adduce the fulfilment of predic

tions as a welcome argument.

In this doctrine, which, from the standpoint of the Stoic doctrine

of the logos, was completely consistent, the opponents of the system

saw of course a decided denial of freedom of the will, and of the

criticisms which the system experienced this was perhaps the

most frequent and at the same time the most incisive. Amongthe numerous attacks the best known is the so-called ignava ratio, or

"lazy reason" (apyos Adyo?), which from the claim of the unavoid-

able necessity of future events draws the fatalistic conclusion that

one should await them inactively, — an attack which Chrysippus

did not know how to avoid except by the aid of very forced distinc-

tions. 2 The Stoics, on the contrary, concerned themselves to showthat in spite of this determinism, and rather exactly by virtue of it,

man remains the cause of his actions in the sense that he is to be

made responsible for them. On the basis of a distinction 3 between

main and accessory causes (which, moreover, reminds us throughout

of the Platonic aiViov and £wamov) Chrysippus showed that every

decision of the will does indeed necessarily follow from the co-opera-

tion of man with his environment, but that just here the outer

circumstances are only the accessory causes, while the assent pro-

ceeding from the personality is the main cause, and to this account-

ability applies. While, however, this voluntarily acting -^ye/xoviKov,

or ruling faculty of man, is determined from the universal Pneuma,this Pneuma takes on in every separate being a self-subsistent

1 Cic. De Fato, 10, 20. So far as concerns disjunctive propositions Epicurusalso for this reason gave up the truth of disjunction ; Cic. De Nat. Deor. I.

25, 70.2 Cic. De Fato, 12, 28 ff.

8 Cic. De Fato, 10, 30 ff.

Page 214: Windelband, History of Philosophy, Vol. 1

194 Hellenistic-Roman Thought : Ethical Period. [Part IL

nature, different from that of others, and this is to be regarded as a

proper ap-^- 1 In particular, the Stoics make prominent the point

that responsibility, as a judgment pronounced on the ethical quality

of actions and characters, is quite independent of the question

whether the persons or deeds might, in the course of events, have

been other than they were, or not.2

4. The problem of the freedom of the will, which had been

already complicated ethically and psychologically, experienced in

this way still further a metaphysical and (in the Stoic sense) theo-

logical complication, and the consequence was that the indeterminists

who were opponents of the Stoa gave a new turn to the conception

of freedom which they regarded as threatened by the Stoic doctrine,

and brought it into sharp definition. The assumption of the excep-

tionless causal nexus to which even the functions of the will were

to be subordinated, seemed to exclude the capacity of free decision;

but this freedom of choice had, since Aristotle, been regarded in all

schools as the indispensable presupposition of ethical accountability.

On this account the opponents thought— and this gave the contro-

versy its especial violence — that they were defending an ethical

good when they combated the Stoic doctrine of fate, and with that

the Democritic principle of natural necessity. And if Chrysippus

had appealed to the principle of sufficient reason to establish this,

Carneades, to whom the freedom of the will was an incontestable

fact, did not fear to draw in question the universal and invariable

validity of this principle.3

Epicurus went still farther. He found the Stoic determinism so

irreconcilable with the wise man's self-determination which formed

the essential feature of his ethical ideal, that he would rather still

assume the illusory ideas of religion than believe in such a slavery

of the soul. 4 Therefore he, too, denied the universal validity of the

causal law and subsumed freedom together with chance under the

conception of uncaused occurrence. Thus in opposition to Stoic

determinism, the metaphysical conception of freedom arose, by means

of which Epicurus put the uncaused function of the will in manupon a parallel with the causeless deviation of the atoms from their

line of fall (cf. § 15, 4). The freedom of indeterminism means,

accordingly, a choice between different possibilities tha,t is deter-

mined by no causes, and Epicurus thought thereby to rescue moral

responsibility.

This metaphysical conception of freedom as causelessness is not at

1 Alex. Aphr. De Fato, p. 112. 3 Cic. De Fato, 5, 9 ; 11, 23 ; 14, 31.2 lb. p. 106. * Diog. Laert. X. 133 f . ; Us. p. 65.

Page 215: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 16.] Physico-Tlieology : Epicurus, Stoics. 195

all isolated in the scientific thought of antiquity. Only the Stoa

held fast inviolably to the principle of causality. Even Aristotle

had not followed into details the application of his general principles

(cf. p. 143); he had contented himself with the eVt to ttoXv, " for the

most part," and had based his renunciation of the attempt fully to

comprehend the particular upon the assumption of the contingent

in Nature, i.e. of the lawless and causeless. In this respect the

Stoics alone are to be regarded as forerunners of the modern study

of Nature.

5. Stoicism encountered difficulties which were no less great, in

carrying out its teleology. The pantheistic system which regarded

the whole world as the living product of a divine Reason acting

according to ends, and found in this its sole ground of explanation,

must of course maintain also the purposiveness, goodness, and perfec-

tion of this universe ; and conversely the Stoics were accustomed to

prove the existence of the gods and of Providence by pointing to

the purposiveness, beauty, and perfection of the world ; that is, by

the so-called pJiysico-theological method. 1

The attacks which this line of thought experienced in antiquity

were directed not so much against the correctness of the reasoning

(though Carneades applied his criticism at this point also) as

against the premises ; and conversely, the easy exhibition of the

many defects and maladaptations, of the evils and the ethical harm

in the world was employed as a counter-reason against the assump-

tion of a rational, purposeful World-cause and of a Providence.

This was done first and with full energy, naturally, by Epicurus,

who asked whether God would remove the evil in the world but

could not, or could remove it but would not, or whether perhaps

neither of these was true,2— and who also pointed to the instances

of injustice in which the course of life so often makes the good

miserable and the wicked happy.3

These objections, intensified and carried out with especial care,

were brought into the field by Carneades.* But to the reference to

the evil and injustice of the course of events he added the objec-

tion to which the Stoics were most sensitive

:

5 " Whence then in

this world which has been created by Reason comes that which is

void of reason and contrary to reason, whence in this world ani-

mated by the divine Spirit come sin and folly, the greatest of all

1 Cic. De Nat. Dear. II. 5, 13 ff.

2 Lactnnt. De Ira Dei, 18, 19 ; Us. Fr. 374.

3 Id. Inst. Div. III. 17, 8 ; Us. Fr. 370.

* Cic. Acad. LI. 38, 120 ; De Nat. Dear. III. 32, 80 ff.

s Cic. De Nat. Dear. III. 25-31.

Page 216: Windelband, History of Philosophy, Vol. 1

196 Hellenistic-Roman Thought : Ethical Period. [Part IL

evils? " And if the Stoics, as perhaps occurred in spite of their

determinism, 1 wished to make free will responsible for these things,

the further question arose, why the almighty World-reason should

have given man a freedom which was thus to be abused, and whyit should permit this abuse.

6. In the presence of such questions the Stoics with their monis-

tic metaphysics were in a much worse case than Plato and Aristotle,

who had been able to trace the maladaptations and evil back to

the resistance of the " Not-being," or of matter respectively. In

spite of this the Stoics came forward boMly to master these diffi-

culties, and brought to light, not without acute thought, most of

those arguments in which at later periods theodicy has moved again

and again.

The teleological doctrine of the perfection of the universe can be

protected against such attacks either by denying the dys-teleological

facts, or by justifying them as the indispensable means or attend-

ant result in the purposefully connected whole. Both methods

were pursued by the Stoa.

Their psychological and ethical theories permitted the claim that

what is called a physical evil is not such in itself, but becomes such

by man's assent, that hence, if diseases and the like are brought

about by the necessity of the natural course of events, it is only

man's fault that makes an evil out of them;just as it is frequently

only the wrong use which the foolish man makes of things that

makes these injurious, 2 while in themselves they are either indif-

ferent or even beneficial. So the objection based on the injustice

of the course of the world is rebutted by the claim that in truth for

the good man and the wise man physical evils are no evils at all,

and that for the bad man, on the other hand, only a sensuous illu-

sory satisfaction is possible, which does not make him truly happy,

but rather only aggravates and strengthens the moral disease which

has laid hold of him.3

On the other hand, physical evils may also be defended on the

ground that they are the inevitable consequences of arrangements

of Nature which are in themselves adapted to their ends and do

not fail of their purpose,— as Chrysippus, for example, attempted to

show in the case of diseases. 4 In particular, however, they have

the moral significance of serving partly as reformatory punishments

of Providence

;

5 partly, also, as a useful stimulus for the exercise

of our moral powers. 6

1 Cleanth. Hymn. v. 17. * Gell. N. A. VII. 1, 7 ff.

2 Seneca, Qu. Nat. V. 18, 4. 5 Plut. Stoic. Rep. 35, 1.8 Seneca, Ep. 87, 11 ff. 6 Marc. Aurel. VIII. 35.

Page 217: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 17. J Criteria of Truth: Peripatetics. 197

While external evils were thus justified principally by pointing

out their ethical purposiveness, it appeared for the Stoics an all the

more urgent problem, though one which proved also the more diffi-

cult, to make moral evil or sin comprehensible. Here the negative

way of escape was quite impossible, for the reality of baseness in

the case of the great majority of men was the favourite subject of

declamation in the Stoic discourses on morals. Here, then, was the

centre of the whole theodicy, namely, to show how in this world

which is the product of divine Reason, that which is contrary to

reason in the impulses, dispositions, and actions of rationally

endowed beings is possible. Here, therefore, the Stoics resorted to

universal considerations. They showed how the perfection of the

whole not only does not include that of all the individual parts, but

even excludes it,1 and in this way substantiated their claim that

God must necessarily allow even the imperfection and baseness of

man. In particular, they emphasised the point that it is only

through opposition to evil that good as such is brought about ; for

were there no sin and folly, there would be no virtue and wisdom. 2

And while vice is thus deduced as the necessary foil for the good,

the Stoics give as a final consideration, 3 that the eternal Providence

ultimately turns even the evil to good, and has in it but an appar-

ently refractory means for the fulfilment of its own highest ends. 4

§ 17. The Criteria of Truth.

The philosophical achievements of the post-Aristotelian time

were least important in the department of logic. Such a powerful

creation as the Analytics of the Stagirite, which brought the prin-

ciples of Greek science in so masterly a fashion to the consciousness

of all in a conclusive form, must naturally rule logical thought for

a long time, and, in fact, did this until the close of the Middle

Ages, and even beyond. The foundations of this system were so

firmly laid that at first nothing there was shaken, and there re-

mained for the activity of the schools but to build up individual

parts, — an activity in connection with which, even at that time,

much of the artificial adornment characteristic of a degenerate age

displayed itself.

1. The Peripatetics had already attempted to develop *-he Aristote-

lian Analytics systematically in this direction by a more detailed treat-

ment, by partially new proofs, by farther subdivision, and by more

1 Plut. Stoic. Rep. 44, G.

2 Ib. 30, 1.

s lb. 35, a4 Cleanth- tLjmn. vv. 18 f.

Page 218: Windelband, History of Philosophy, Vol. 1

198 Hellenistic-Roman Thought : Ethical Period. [Part IL

methodical formulation. In particular, Eudemus and TJieophrastus

undertook investigations concerning the hypothetical and disjunc-

tive judgments, and the extension of the theory of the syllogism

occasioned by the appearance of these judgments and premises.

The Stoics continued these efforts ; they set these new forms of

judgment (d£tw/Aa) as composite over against the simple 1 categorical

forms, developed into all their details the resulting forms of the

syllogism, emphasised also especially the quality z of judgments,

and deduced the laws of thought in altered forms. In general,

however, they spun out the logical rules into a dry schematism and

genuine scholastic formalism which thereby became farther and

farther removed from the significant fundamental thoughts of the

Aristotelian Analytics, and became a dead mass of formulae. The

unfruitful subtlety of this process took special delight in the solu-

tion of sophistical catches, in which the real meaning was inextri-

cably involved in the contradiction of forms.

It was in these elaborations by the schools that the science of

logic created by Aristotle first took on the purely formal character

that it retained up to the time of Kant. The more pedantic the

form taken in the development of the particular features, the more

the consciousness of the living thought, to which Aristotle had

aspired, was replaced by a schoolmaster-like network of rules,—essentially designed to catch thoughts and examine their formal

legitimacy, but incapable of doing justice to the creative power of

scientific activity. While, even with Aristotle, regard for proof

and refutation had occupied the foreground, here it occupies the

whole field. Antiquity did not attain a theory of investigation

;

for the weak beginnings which we find toward this end in the inves-

tigations of a younger Epicurean,3 Philodemus,4 concerning conclu-

sions from induction and analogy, are relatively isolated, and have

no result worthy of mention.

2. In the doctrine of the Categories, of the elaboration of which the

Stoics made much account, more that was real was to be expected.

Here it was indeed quite correct, and yet not very fruitful, to call

attention to the fact that the supreme category, of which the rest

1 Sext. Emp. Adv. Math. VIII. 93.2 Diog. Laert. VII. 65.3 Epicurus himself, and his school also, as a whole, did not trouble themselves

as to the principles of formal logic. One might regard this as an evidence of

taste and intelligence, but it was in truth only indifference toward all that did

not promise directly practical advantages.4 On his treatise wepl o-qiidwv Kai ffypeidxretav, discovered in Herculaneum, cf.

Th. Gompertz, Herculanensisehe Studien, Heft 1 (Leips. 1865) ; Fr. Bahusch(Lyck, 1879); R. Philmpson (Berlin, 1881).

Page 219: Windelband, History of Philosophy, Vol. 1

Chap. 1, §17.] Criteria of Truth : Stoics. 199

represent only special determinations, is that of Being (tow) 1 or

Something (t2); and the co-ordination of the categories winch, at

least as regards the method of their enumeration, was Aristotle's

plan, was replaced by an expressly systematic succession, according

to which each category was to be more exactly determined by the

following one. " What is," or Being, as abiding substrate of all

possible relations, is substance (u7ro/cei/x£vov) ; this is the supporter

{Träger) of fixed qualities (71-010V), and only in this aspect is it

involved in changing states (to 7rws «^ov), and, in consequence of

these latter, in relations to other substances (to 7rpos rt 71-cos *xov )-

Out of the doctrine of the categories grows thus an ontology, that

is, a metaphysical theory as to the most general formal relations of

reality, and this theory in the system of the Stoics, agreeably to

their general tendency (cf. § 15, 5), takes on a thoroughly materi-

alistic character. As substance, the existent is matter which is in

itself destitute of properties (v\rj), and the qualities and forces

which are inherent in matter as a whole, as well as in a particular

part (TroioT-qrts — Swa/Lieis) , are likewise kinds of matter (atmospheric

currents) which are commingled with it («paons St' oXwv). In this

connection both substance and attributes are regarded, as well from

the point of view of the general conception as from that of the indi-

vidual thing, and in the latter aspect it is emphasised that every

individual thing is essentially and definitely distinguished from all

others. 2

Besides these categories of Being, we find making their appear-

ance among the Stoics those conceptional forms by which the rela-

tion of thought to Being is expressed, and in these the separation of

the subjective from the objective, for which a preparation had been

growing more and more complete in the development of Greek

thought, now attains definite expression. For while the Stoics

regarded all objects to which thought relates as corporeal, while

they regarded the activity of thought itself, and no less its expres-

sion in language 3 as corporeal functions, they were still obliged to

confess that the content of consciousness as such (to Xektov) is of in-

1 That the Peripatetics also husied themselves with this category is provedhy the definition preserved by Strato : to 5v e<m tö ttjs Sia^o^s atnov (Proclus

in Tim. 242 E).- In contrasting the first two with the last two categories, the language rela-

tion of noun and verb appears here also (in Stoic terminology ttt&ctis and kotij-

y6pijßa).3 The Stoics laid great weight upon the discriminative comparison of thought

and of speech, of the inner activity of reason (A670S ivdiaQeros), and of its ex-

pression through the voice (\670s Trpo^optKÖs) . Hence, too, the assumption (cf.

§ 15, (1) of the faculty of speech as a proper part of the soul;hence their thor-

ough treatment of rhetoric and grammar side by Side with logic.

Page 220: Windelband, History of Philosophy, Vol. 1

200 Hellenistic-Roman Thought : Ethical Period. [Part II

corporeal nature. But since the distinction was thus sharply

drawn between Being and content of consciousness, the fundamental

epistemological problem came forward, how the relations by which the

ideational content refers to Being and agrees with it, are to be thought.

3. This question was, moreover, also brought home by the vigor-

ous development which Scepticism had meanwhile undergone, and

by the relatively strong position which it occupied as compared with

the dogmatic systems.

Whether by Pyrrho or Timon it matters not, it was at all events

at about the same time at which the great school-systems became

dogmatically developed and fortified, that all those arguments were

systematised into a complete whole, by which the Sophistic period

had shaken the naive trust in man's capacity for knowledge. Al-

though the ethical end of making man independent of fate by with-

holding judgment was ultimately decisive (cf. § 14, 2), this

Scepticism still forms a carefully carried out theoretical doctrine.

It doubts the possibility of knowledge in both its forms, the forrr

of perception as truly as that of judging thought, and after it has

destructively analysed each of these two factors singly, it adds

expressly that just on this account their union can have no certain

result. 1

As regards perception, the Sceptics availed themselves of the

Protagorean relativism, and in the so-called ten Tropes 2 in which

yEnesidemus 3 sets forth the sceptical theory with very defective

arrangement, this tendency still occupies the broadest space. Per-

ceptions change not only with the different species of animate

beings (1), not only with different men (2), according to their cus-

toms (9) and their whole development (10), but even in the case

of the same individual at different times (3), in dependence upon

bodily conditions (4), and upon the different relations in which the

individual finds himself with regard to his object spatially (5).

They alter, also, because of the difference in the states of the object

(7), and have, therefore, no claim to the value of an immediate

report of things, because their origination is conditioned by inter-

mediate states in media such as the air, the co-operating elements

furnished by Which we are not able to deduct (6). Man is, there-

1 From two deceivers combined it is only right to expect no truth. Diog,

Laert. IX. 114.2 Sext. Emp. Pyrrh. Hyp. I. 38 ff.

3 It was said by the ancient writers that iEnesidemus was attached, not onlyto Scepticism, but also to the metaphysics of Heraclitus. The question whetherthis was actually so, or whether such a relation was only ascribed to him by mis-take, has solely antiquarian significance. For had the former been the case, it

would have been but another manifestation of a real relationship in thought, to

Which Plato had already directed attention. Thecet. 152 E ff. ; cf. p. 92, note 2.

Page 221: Windelband, History of Philosophy, Vol. 1

CHAr. 1, § 17.] Criteria of Truth: Sceptics. 201

fore, in all ways, not in a condition to know things purely (8), and

in the face of the multiplicity of impressions so full of contradic-

tions he has no means of distinguishing a true from a false impres-

sion. One is no more (ov tioAJW) valid than another.

Equally relative with man's perceptions are also his opinions

(So£ai). In this aspect the influences of the Eleatic dialectic

assert themselves in Pyrrhonism. It is shown that to every opinion

the opposite can be opposed with equally good reasons, and this

equilibrium, of reasons {la-oa-Oivua tS>v \6ywv) does not permit us,

therefore, to distinguish true and false : in the case of such a con-

tradiction (diT<Aoy6a) the one holds no more than the other. All

opinions accordingly stand— according to the phrase of the

Sophists, adopted by the Sceptics— only by convention and cus-

tom (vo/xü) re Kai !0e<.), not by their essential right and title (<£ucm).

More energetically still did the later Scepticism attack the possi-

bility of scientific knowledge, by disclosing the difficulties of the

syllogistic procedure, and of the methods which Aristotle had built

up upon this. 1 In this Carneades seems to have led the way, show-

ing that every proof, since it presupposes other proofs for the valid-

ity of its premises, makes necessary a regressus in infinitum— an

argument that was completely in place for the Sceptic who did not,

as did Aristotle, recognise anything as immediately certain (a/xeo-ov

;

cf. § 12, 4). The same argument was carried further by Agrippa,

who formulated Scepticism in five Tropes 2 much more clearly and

comprehensively than iEnesidemus. He called attention again to the

relativity of perceptions (3) and of opinions (1) ; he showed howevery proof pushes on into infinity (2 : 6 eis anupov iKßdWuiv), and howunjustifiable it is in the process of proof to proceed from premises

that are only hypothetically to be assumed (4), and finally, howoften it occurs, even in science, that that must be postulated as

ground of the premises which is only to be proved by means of the

syllogism in question (5: 6 SiaAA^Aos). In the latter aspect atten-

tion was also called to the fact that in the syllogistic deduction of

a particular proposition from a general one, the general would yet

from the outset be justified only on condition that the particular

were valid. 3

Since the essential nature of things is thus inaccessible to human

i Sext. Emp. Adv. Math. VIII. 316 ff.

2 Sext. Emp. Pyrrh. Hyp. I. 164 ff. : (1) The conflict of opinions. (2) Theendless regress in proving. (3) The relativity of all perceptions. (4) The im-possibility of other than hypothetical premises. (5) The circle in the syllogism.

8 Sext. Emp. Pyrrh. Hyp. II. 104 ff. Renewed in J. S. Mill, Logic, II. 3, 2;

corrected iu Chr. Sigwart, Logik, I. § 55, 3.

Page 222: Windelband, History of Philosophy, Vol. 1

202 Hellenistic-Roman Thought : Ethical Period. [Part II.

knowledge, 1 the Sceptics demanded that man should suspend judg-

ment so far as possible (iiroxv)- We can say nothing concerning

things (a<f>a<Tta) ; we can only assert that this and that appears so

or so, and in so doing we report only our own momentary states (as

the Cyrenaics had already taught, § 8, 3). Even the sceptical main-

tenance of the impossibility of knowledge (in order to avoid the

contradiction that here something of a negative character, at least,

seems to be maintained and proved) 2 should be conceived of rather

as a profession of belief than as knowledge,—more as a withholding

of opinion than as a positive assertion.

Cf. V. Brochard, Les Sceptiques G-recs (Paris, 1877).

4. The attack of Scepticism was most sharply concentrated in

the principle 3 that, in the presence of the deceptions to which

man is exposed in all his ideas of whatever origin, there is no uni-

vocal, sure sign of knowledge, no criterion of truth. If, therefore,

the dogmatic schools held fast to the reality of knowledge, even

from the Socratic motive that virtue is impossible without knowl-

edge, 4 they found the task assigned them by this sceptical position

of announcing such a criterion and of defending it against the

sceptical objections. This was done also by the Epicureans and

Stoics, although their materialistic metaphysics and the sensualistic

psychology connected with it prepared for them serious, and,

ultimately, insurmountable difficulties.

In fact, it was the psycho-genetic doctrine of both these schools

that the content of all ideas and knowledge arises solely from sen-

suous perception. The origin of sense-perception the Epicureans

explained by the image theory of Democritus (§ 10, 3). This

theory gave even to the illusions of the senses, to dreams, etc., the

character of perceptions corresponding to reality ; and even the con-

structions of the combining fancy or imagination could be explained

on this theory by unions which had already taken place objectively

between the images. But the Stoics also regarded perception as a

bodily process, as an impression of outer things upon the soul

(rviraxns) , the possibility of which seemed to them to be self-

evident, in view of the universal commingling of all bodies. This

1 The simplest formulation of Scepticism, finally, was that which broughtAgrippa's five Tropes together into two ; there is nothing immediately certain,

and just on this account nothing mediately certain; accordingly nothing what-ever that is certain. Sext. Emp. Pi/rrh. Hyp. I. 178 f.

2 Cic. Acad. II. 9, 28 and 34, 109'; Sext. Emp. Adv. Math. VIII. 463 ft3 Sext. Emp. Adv. Math. VII. 159.4 Diog. Laert. X. 146 f . K. A ; Us. p. 76 f., on the other hand, Plut Stoic.

Hep. 47, 12.

Page 223: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 17.] Criteria of Truth : Epicureans, Stoics. 203

erassly sensuous conception they expressed by the since frequently

repeated comparison, that the soul is originally like a blank tablet,

on which the outer world imprints its signs in the course of time. 1

More refined, but more indefinite, and yet absolutely mechanical

still in its tone is the designation of Cbrysippus, who called percep-

tion an alteration of qualities (erepoiWis) in the soul ; for, at all

events, the idea or mental presentation (^»avrao-ia) remains for

him, too, a corporeal effect or product of that which is presented

(<f>avTa<TTOv) .

Both schools explained the presence of conceptions and of general

ideas (tt/doX^i/kis, and among the Stoics also kolvoX Zwoiai) solely by

the persistence of these impressions, or of parts of them, and by

their combination. They combated, therefore, as the Cynics espe-

cially had already done, the Platonic-Aristotelian doctrine of Ideas

and Forms, 2 especially the assumption of an independent activity

or power of forming conceptions, and traced even the most general

and abstract conceptions back to this mechanism of elementary

perceptions (to which they scarcely gave any further analysis).

To these general ideas of experience (i/xTeLpca), which arise natu-

rally and involuntarily (^vo-tKws) , the Stoics indeed opposed the

conceptions of science produced by the aid of a methodical con-

sciousness ; but even the content of these scientific conceptions was

held to be exclusively derived from sensations. In this connection,

both schools laid especial weight upon the co-operation of language

in the origination of conceptions.

But now, in so far as the total content of impressions, and like-

wise also the nature of thought, are the same among all men, it

necessarily follows that under these circumstances the same general

ideas will be formed, in both the theoretical and the practical domain,

by means of the psychological mechanism. This consequence was

drawn especially by the Stoics, whose attention was by their whole

metaphysics directed vigorously to the common nature of the psy-

chical functions, which were all held to arise from the divine Pneuma.

They taught, therefore, that the surest truth is to be sought in those

ideas which develop uniformly among all men with natural neces-

sity, and they liked to take as their starting-point, even for scientific

reasonings, these Koivai evvocu, or communes notiones. They have a

1 Plut. Plac. IV. 11 ; Dox. D. 400 ; Plut. Comm. Xot. 47 ; cf. besides Plat.

Thecet. 191 C.2 Hence the Stoics regard Platonic " Ideas " (class-concepts) as merely struc-

tures of the human mind (iwormaTa rjnirepa ; cf. Plut. Plac. 1. 10, Dax. D. 309),and thus gave the first suggestion for the later subjective meaning of the term"idea." Cf. § 19.

Page 224: Windelband, History of Philosophy, Vol. 1

204 Hellenistic-Roman Thought : Ethical Period. [Part II.

predilection for appealing to the consensus gentium— the consent of

all men, — an argument whose validity it was easy for the Sceptics

to shake by pointing to the negative instances of experience. 1

It was, therefore, not in the spirit of the Stoics that in the later

Eclectic literature these common ideas were called innate (innatce),

and that Cicero especially saw in them not only that which Nature

teaches equally to all, but also that which Nature or the deity has

originally implanted in every one at the same time with his reason.

Cicero maintains this, not only for the fundamental conceptions of

morality and right, but also for the belief in the deity and in the

immortality of the soul : the knowledge of God especially is held

to be only man's recollection of his true origin. 2 This doctrine

formed the best bridge between the Platonic and the Stoic theories

of knowledge, and under the Stoic name of Koival Zwoai the ration-

alistic doctrine of knowledge was propagated on into the beginnings

of modern philosophy. Just by this means it retained the accessory

psychologists meaning that rational knowledge consists in innate ideas.

5. While now the Stoics as well as the Epicureans originally

traced back all the contents of ideas to sense-impressions psycho-

genetically, it was only the Epicureans who drew from this the

consistent inference that the sign for the recognition of truth is

solely the feeling of the necessity with which a perception forces

itself upon consciousness, the irresistible clearness or vividness

(ivdpyaa) conjoined with the taking up of reality in the function

of the senses. Every perception is as such true and irrefutable ; it

exists, so to speak, as a self-certain atom of the world of conscious-

ness, free from doubt, independent, and unmovable by any reasons

whatever. 3 And if different and mutually contradictory perceptions

of the same objects seem to exist, the error lies only in' the opinion

which refers them, and not in the perceptions which by the very

fact of their difference prove that different outer causes correspond

to them ; relativity is accordingly nothing in point against the cor-

rectness of all perceptions. 4

Meanwhile, opinions (So&u) constantly and necessarily go beyond

this immediate presence of sense-impressions : for the knowledge

requisite for acting needs also knowledge of that which is not

immediately perceptible : it needs to know, on the one hand, grounds

i Cic. De Nat. Deor. I. 23, 62 f.

2 Id. De Leg. I. 8, 24 : ... ut is agnoscat deum, qui unde ortus sit quasi re-

cordetur ac noscat.3 The parallelism of this episternological Atomism with the physical and

ethical Atomism of the Epicureans is obvious.* Sext. Emp. Adv. Math. VII. 203 ff.

Page 225: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 17.] Criteria of Truth : Epicureans. 205

of phenomena (a8i;Xov), and on the other hand the expectation as to

the future that may be inferred from them (irpoa-^ivov). But for all

these farther functions of the psychical mechanism there is, accord-

ing to the Epicureans, no other guaranty than perception again.

For if conceptions (7rpo\^if/ets) are only sense-impressions retained

in the memory, they have their own certainty in the clearness or

vividness of these impressions, a certainty susceptible neither of

proof nor of attack; x and hypotheses (woA^i/rcis), both with regard

to the imperceptible grounds of things and also with regard to future

events, find their criterion solely in perception, in so far as they are

verified by it, or at least not refuted ; the former holds for the pre-

diction of the future, the latter for explanatory theories. 2 There

is therefore among the Epicureans nothing said of an independent

faculty of conviction or belief ; whether our expectation of any event

is correct we can know only when the event occurs. Thus they re-

nounce on principle any attempt at an actual theory of investigation.

6. It is evident from this that the Epicureans might regard their

own Atomistic metaphysics as a hypothesis not refuted by facts, but

that they were not permitted to regard it as a hypothesis that was

proved. It was a hypothesis, indeed, of which the essential end, as

they employed it, was to displace other hypotheses which seemed

to them ethically objectionable. Their dogmatism is accordingly

only problematical, and their doctrine of knowledge, in so far as it

has to do with rational knowledge, is very strongly permeated with

scepticism. In so far as they recognise only that which passes with

sense-perception as a " fact," but regard such facts as completely cer-

tain, their standpoint is to be designated as that of Positivism.

This positivism was developed in antiquity still more consistently,

and in a form freed from the ethical and metaphysical tendencies of

Epicurus, by the theories of the later schools of empirical physi-

cians. These schools went with the Sceptics as regards knowledge

of all that is imperceptible by the senses and as regards all rational

theories ; on the other hand, in their recognition of the sensuous

evidence of perceptions, they went with the Epicureans. Observation

(tt/pt/o-is) is here portrayed as the basis of the physician's art, and ob-

servation retained in memory is regarded as the sole essence of his

theory : aetiological explanations especially are rejected on principle.

Connected with this is the circumstance that the later Sceptics

treated the conception of causality in searching investigations and

1 As the final criterion even for the intellectually good is, with Epicurus, sen-

suous pleasure, so the criterion of the truth of conceptions is only sensuousvividness (Evidenz).

2 Sext. Emp. VII. 211.

Page 226: Windelband, History of Philosophy, Vol. 1

206 Hellenistic-Roman Thought : Ethical Period. [Part II.

discovered its difficulties. iEnesidemus had already propounded a

series of such aporiae, 1 and in Sextus Empiricus we find thern devel-

oped more broadly and comprehensively. 2 With him not only such

defects of serological theories are designated as, that they reduce

the known to the unknown whicn is just as inexplicable, that they

maintain one possibility among many without a sufficient reason,

that they do not examine experience carefully enough with a view

to possible negative instances, and finally that they after all explain

that which is inaccessible to perception by some sort of a scheme

known from perception, which is especially simple and therefore

apparently intelligible in itself ; besides these, he searches out, also,

all the general difficulties which prevent us from gaining a clear

(picturate) idea of the causal relation. The process of the action

of one thing upon another, the passing over of motion from one

thing to another, can be made intelligible neither on the assumption

that that which acts (as force) is immaterial, nor on the opposite

assumption ; nor does contact (d4»rj) which is assumed as a conditio

sine qua non of the causal process (as had been already done by

Aristotle) make it any more explicable. So, too, the time relation

of cause and effect is extremely difficult to determine. The most

important thought in these discussions, however, is the pointing out

of the relativity of the causal relation : nothing is in itself a cause or

effect; each of the two h such only with reference to the other;

oltiov and 7racrxov are correlative terms which must not be absolutely

postulated or asserted. The (Stoic) conception of an essentially

efficient cause, the conception of a creative deity, is then thereby

excluded.

7. The Sceptics of the Academy sought in another direction a

substitute for the certainty of rational knowledge which they also

had given up. Since in practical life suspense cannot be carried

out as a principle of conduct and action is indispensable, and since

for action determining ideas are requisite, Arcesilaus brought out

the view that ideas, even though one refuse them his complete

assent, are yet able to move the will,3 and that in practical life one

must content himself with a certain kind of confidence or trust

(ttio-tis), according to which some ideas may in a greater degree than

others be regarded as probable (evXoyov), adapted to the purpose of

life, and reasonable.*

1 Sext. Emp. Pyrrh. Hyp. I. 180 ff.

2 Adv. Math. IX. 195 ff. ; cf. K. Göring, Der Begriff der Ursache in der grie-

chischen Philosophie (Leips. 1874).3 Plut. Adv. Col. 26, 3.

* Sext. Emp. Adv. Math. VII. 158.

Page 227: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 17.] Criteria of Truth : Sceptics, Stoics. 207

The theory of Probabilism was carried out farther by Carneades '

in an attempt to define more exactly, according to logical relations,

the particular degrees of this " belief." The least degree of proba-

bility {inOavoTrj'i) is that which (as an indistinct and imperfect form

of sensuous clearness or vividness— ivdpyeta) belongs to the single

idea that stands in no farther connections. A higher degree of

probability belongs to that idea which can be united (dTrepiWuo-Tos),

without any contradictions, with other ideas in connection with

which it belongs. Lastly, the highest stage of belief is reached

where a whole system of such connected ideas is examined as to its

complete harmony and verification in experience (ttc/oiwSeu/acvt;).

Empirical confidence rises, therefore, from the sensuously isolated

to the logical systems of scientific research. But though in the

latter form it may be completely sufficient for practical life (as

Carneades assumed), it is yet not able to lead to a completely

certain conviction.

8. In contrast with this, the Stoics made the most strenuous

efforts to gain an epistemological substructure for their metaphysics,

to which they attributed so high a value from considerations of ethi-

cal interest, and in spite of psycho-genetic sensualism, to rescue the

rational character of science. 2 On the principle that like is known

by like, their doctrine of the World-reason demanded a knowledge

of the external Logos by the internal logos of man,— by his rea-

son;

3 and the ethical antagonism or dualism between virtue and

the sensuous impulses required a parallel distinction between

knowledge and sensuous ideas. Although, therefore, the whole

material of knowledge was held to grow out of sensuous presenta-

tions, the Stoics pointed out, on the other hand, that in perception

as such, no knowledge whatever is contained ; that it is not to

be characterised as either true or false. Truth and falsity can be

predicated only when judgments (a£iw//.<«"a) have been formed in

which something is asserted or denied as to the relation of ideas. 4

Judgment, nevertheless, is conceived of by the Stoics— and in

this they take a new and important position, which, in antiquity,

only the Sceptics approach in some degree— by no means merely as

the theoretical process of ideation and combination of ideas. They

recognised, as the essential characteristic in judgment, the peculiar

act of assent (cruyicaTa0e<n$)> of approval, and of being convinced,

with which the mind makes the content of the idea its own, grasps

1 Ib. J66 ft".

2 Cf. M. Heinze, Zur Erkenntnisslehre der Stoiker (Leips. 1880V8 Sext. Emp. Adv. Math. VII. 93.

* Sext. Emp. Ado. Math. VIII. 10.

Page 228: Windelband, History of Philosophy, Vol. 1

208 Hellenistic-Roman Thought : Ethical Period. [Part IL

it, and in a certain "way takes possession of it {KaTaXafxßdveiv) . This

act of apprehension the Stoics regard as an independent function of

consciousness (^ye^oi/iKoV) , in the same way as they regard the

assent to the impulses, which makes its appearance in passion. The

arising of ideas, like that of the excitations of feeling, is a process

which is of natural necessity and completely independent of humanwill (Slkovo-lov) ; but the assent by which we make the one class,

judgments, and the other, passions, is a decision (k^utis) of con-

sciousness, free (exotxnov) from the outer world.1

But now in the case of the wise man, by virtue of the identity of

the universal with the individual logos, this assent appears only in

the case of those ideas which are true : the soul, therefore, in appre-

hending the content of these ideas, apprehends reality. Such an

idea the Stoics called <f>avraaca KaTakrj-n-TiKrj,2 and they were of the

conviction that such an idea must call forth the reasonable man's

assent with immediate evidence or clearness. Hence assent itself

(o-vyKardOeo-Ls) is conceived of as an activity of the thinking soul,

but individual perceptions appear as the objects of assent as truly

as do the intellectual activities of conception, judgment, and reason-

ing, based upon the individual perceptions.

If thus the Stoics understood by the favracria. KaraX-q-n-rLKrj that

idea by which the mind lays hold of reality, and which, therefore,

so illumines the mind that this, in its assent, makes reality its own,

this was indeed the correct expression for the requirement which

they set up for the true idea, 3 but the definition was not at all

adapted to the end for which it was framed : that is, for a sign by

which to recognise truth. For as the Sceptics 4 very justly objected,

the subjective mark, assent, might be shown as a psychological fact

in the case of a multitude of evidently false ideas.

Thus the anthropological discord in the Stoic doctrine manifests

1 lb. VIII. 39, 7.

2 In the interpretation of this term there is a wide divergence. According to

the sources, it seems now as if the idea were intended which the mind lays holdof, now that which apprehends the real fact, now that by which the mind appre-hends reality, and now again that which on its part so lays hold of the mindthat the mind must assent to it. It has hence been supposed that the Stoics

purposely constructed the expression in this ambiguous form, inasmuch as all

these relations would harmonise in it, and perhaps E. Zeller (IV.3 83) [Eng. tr.,

Stoics, etc., p. 89] intended to repeat this ambiguity by his translation, " concep-tional idea or perception" (begriffliche Vorstellung), which, however, has anaccessory logical sense that the Stoics certainly did not intend.

3 It is worth while to point out the fact that in their designations for therelation of the knowing mind to the external reality, the Stoics employ, for themost part, expressions from the field of the sense of touch (impression, appre-hending, or grasping, etc.), while formerly optical analogies had been preferred.

Cf. § 11, 2.

i Sext. Emp. Adv. Math. VII. 402 ff.

Page 229: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 17.] Criteria of Truth : Stoics. 209

itself even in this central conception of their theory of knowledge.

As it could not be explained in accordance with their metaphysics

how the individual soul arising from the World-reason should fall

under the mastery of sensuous impulses, so it is ecpially impossible

to understand how theoretical assent should, under certain circum-

stances, be given even to false ideas. Both difficulties, however,

have ultimately a common ground. The Stoics agreed with Hera-

clitus in identifying in their metaphysics the normative and the

actual ordering of things, although these conceptions had meanwhile

become much more clearly separated. Reason was for them that

which should be, as well as that which is ; it was at the same time

vo/aos and <f>v(TLs. And this antithesis, the two sides of which came

into strenuous opposition in their doctrine of freedom and their

theodicy, was the problem of the future.

Page 230: Windelband, History of Philosophy, Vol. 1

CHAPTER IL

THE RELIGIOUS PERIOD.

J. Simon, Histoire de VEcole cVAlexandrie. Paris, 1843 ff.

E. Matter, Essai sur VEcole d'Alexandrie. Paris, 1840 ff.

E. Vacherot, Histoire Critique de VEcole d'Alexandrie. Paris, 1846 ff.

[J. Drammond, Philo Judaius, or the Jewish Alexandrian Philosophy in its

Development and Completion. 2 vols., Lond. 1888.]

Barthelemy St. Hilaire, Sur le Concours ouvert par VAcademie, etc., sur VEcole

d'Alexandrie. Paris, 1845.

K. Vogt, Neuplatonismus und Christenthum. Berlin, 1836.

Georgii, Ueber die Gegensätze in der Auffassung der alexandrinischen Beligions-

philosophie (Zeitschr. f. hist. Theol. 1839).

E. Deutiiiger, Geist der christlichen Ueberlieferung. Regensburg, 1850-51.

A. Ritschi, Die Entstehung der altkatholischen Kirche. 2d ed., Bonn, 1857.

Chr. Baur, Das Christenthum der drei ersten Jahrhunderte. Tübingen, 1860.

J. Alzog, Grundriss aer Patrologie. 3d ed., Freiburg i. B. 1876.

[A. V. G. Allen, The Continuity of Christian Thought. Boston, 1884.]

Alb. Stöckl, Geschichte der Philosophie der patristischen Zeit. Würzburg, 1859.

J. Huber, Die Philosophie der Kirchenväter. Munich, 1859.

Fr. Overbeck, Ueber die Anfänge der patristischen Litteratur (Hist. Zeitschr.

1882).

A. Harnack, Lehrbuch der Dogmengeschichte. 3 vols. Freiburg i. B. 1886-90.

[J. Donaldson, Critical History of Christian Literature and Doctrine.}

The gradual transition of the Hellenistic-Roman philosophy from

the ethical to the religious standpoint had its inner causes in this

philosophy itself, and its external occasion in the imperious de-

mands made by the felt need of the time. For the farther the

contact between the systems extended, the more it became evident

how little able philosophy was to fulfil the task which it had set

itself : namely, that of educating man by a sure insight to a state

of virtue and happiness, to inner independence of the world. While

the sceptical mode of thought, which was extending more and more,

already taught that virtue consists rather in the renunciation of the

attempt to know, than in knowledge itself, the view forced its waymore and more, even among the Stoics, that their ideal of the wise

man, so sharply and rigidly drawn, was not entirely realised in any

210

Page 231: Windelband, History of Philosophy, Vol. 1

Chap. 2.] The lieligious Period. 211

human being, and thus it was felt in every direction that man in

his own strength can become neither knowing, nor virtuous andhappy.

If, then, a disposition to welcome a higher help for ethical ends

was necessarily evoked in philosophy itself, it was also true that

the theoretical doctrines of the time contained a great number of

religious elements. The Epicureans, to be sure, purposely excluded

such, but the Stoics, on the contrary, granted them an entrance that

was all the freer. With the Stoics, not only did metaphysics lead

to seeking the principle of morals in a divine command, but in their

pnmma doctrine, the possibility presented itself of giving to the

creations of myth a philosophical meaning, which might be shared

also by all forms of worship. Finally, the spiritual monotheism in

Aristotle's teaching, and that ideal tendency with which Plato

sought the abiding essence of things in a higher world of the super-

sensuous, were not forgotten.

Just this dualism, which opposed the earthly world of the perish-

able to a supersensuous world of the divine, ultimately proved to be

the right expression for that inner discord which ran through the

entire life of the aging Greek and Roman world. The old craving

for sensuous pleasure might still celebrate its orgies in full powerand to the intoxication of the senses ; but in the midst of it all, out

of surfeit and loathing grew a new craving for a purer, higher joy :

and in the presence of the tremendous contrasts which the social

condition of the Roman Empire brought with it, the look of all the

millions that saw themselves excluded from the good things of this

earth turned longingly toward a better world. Thus in all ways a

deep, passionate need for true salvation of the soul (o-cor^/Dia) cameto be increasingly felt, a hunger for something beyond the earthly,

a religious urgency without an equal.

This religious movement proved its vigour first of all in the eager

reception which foreign forms of worship found in the Graeco-

Roman world, in the mingling and fusing of Oriental and Occidental

religions. But with the adjustment which their oppositions found

here and there, their strife for the mastery over men's spirits be-

came still more energetic, and thus the soil of the ancient world of

civilisation, after bearing the fruits of art and science, became the

battleground of religions. Man's essential interest became thereby

transferred for long centuries from the earthly to the heavenly

sphere; he began to seek his salvation beyond the world of sense.

But the forms in which this contest of the religions was wagedprove in spite of all what a spiritual and intellectual power Greekscience had grown to be. For so strongly was the ancient world

Page 232: Windelband, History of Philosophy, Vol. 1

212 HeUeiiistic-Roman Thought. [Pakt IL

"sicklied o'er with the pale cast of thought," so deeply had it be-

come permeated by the feeling of a need for knowledge, that each of

the religions desired to satisfy not only the feelings but also the

intellect, and was therefore anxious to transform its life into a doc-

trine. This is true even of Christianity, and indeed precisely true of

it. The true, victorious power of the religion of Jesus lay, to be

sure, in the fact that it entered this decrepit, blase world with the

youthful force of a pure, high, religious feeling, and a conviction

that was courageous to the death ; but it was able to conquer the

ancient civilised world only by taking it up into itself and working

it over ; and as in its external conflict with the old world it shaped

its own constitution : and thereby ultimately became so strong as to

be able to take possession of the Eoman state, so also in its defence

against the ancient philosophy it made the world of that philoso-

phy's ideas its own, in order thereby to build up its own dogmatic

system.

Thus the needs of science and of life met. The former sought the

solution of the problems at which it had been labouring in vain, in

religion, and the latter desired a scientific formulation and basis for

its religious longing or conviction. Hence from this time on, for

many centuries, the history of philosophy is grown together with

that of dogmatic theology,2 and the period of religious metaphysics

begins. The thought of antiquity described a peculiar curve, sepa-

rating itself farther and farther from religion from which it pro-

ceeded, reaching its extreme separation in Epicureanism, and then

again steadily drawing near to religion, to return at last entirely

within it.

Under these conditions it is possible to understand how that

Weltanschauung which separated the supersensuous and the sensu-

ous,— looking upon them, from the point of view of value, as divine

perfection and earthly baseness, respectively,— constituted the

common ground of the whole religious-philosophical movement.

This view had already, indeed, been introduced by the Pythagoreans

(cf. § 5, 7), and had been maintained even by Aristotle, but it had,

without doubt, found its most forcible formulation in the Platonic

metaphysics. It was, therefore, this latter system which formed the

controlling centre for the religious closing development of ancient

thought. A religious development of Platonism is the fundamental

character of this period.

1 Cf. K. J. Neumann, Der römische Staat und die allgemeine Kirche bis aufDiocletian (Vol.I. Leips. 1890).

2 It will be understood as a matter of course that the following exposition

has left at one side all specifically dogmatic elements, except where they are

quite inseparably interwoven with philosophical principles.

Page 233: Windelband, History of Philosophy, Vol. 1

chap. 2.] The Religious Period. 213

The geographical centre of the movement, however, is found in

that city which, by its history, as well as by its population, repre-

sented most distinctly the mingling of peoples and of religions,—Alexandria. Here, where in the active work of the museum all

treasures of Grecian culture were garnered, all religions and forms

of worship crowded together in the great throngs of the commercial

metropolis to seek a scientific clarification of the feelings that

surged and stormed within them.

The first line of the Alexandrian philosophy is the so-called Neo-

Pythagoreanism, a mode of thought which, proceeding from the

religious practice of the Pythagorean mysteries, makes only an

external use of the number-mysticism of the old Pythagoreans after

whom it calls itself and its writings, while it finds the theoretical

setting for its world-renouncing, religious-ascetic ethics in a trans-

formation of the Platonic metaphysics, which became of the pro-

foundest value for the conception of the spiritual nature in the

following period. Apollonius of Tyana, the founder of a religion,

is to be regarded as typical representative of this school.

Not without influence from this school, the Stoa, also, in the time

of the Empire, brought out more energetically the religious elements

in its theory of the world, so that not only did the anthropological

dualism of the system become sharpened, but a more theistic mode

of thought gradually became substituted for the original pantheism

of the school. In men like Seneca, Epictetus, and Marcus Aurelius,

the Stoic doctrine became completely a philosophy of deliverance or

redemption.

Even Cynicism revived again about this time in a religious garb,

as a rude, popular preaching of renunciation, and Demonax passes

for its best-known representative.

Scarcely to be separated from the Neo-Pythagoreans are the

Eclectic Platonists of the first centuries of our era, such as Plutarch of

Chseronea and Apideius of Madaura. Later appear Numenius of

Apamea and Nicomaclms of G-erasa, who, besides, already stand

under Jewish and Christian influences as witnesses of a complete

fusion of the two tendencies.

But while, in all these forms, the Hellenic element ever maintains

the ascendency over the Oriental, the latter makes its appearance in

very much stronger force in the Jewish philosophy of religio)). As

the sect of the Essenes * probably proceeded from a contact of Xeo-

Pythagoreanism with the Hebrew religious life, so the various

attempts of learned Jews to draw nearer to Greek science in the

1 Cf. E. Zeller V.3 277 ff.

Page 234: Windelband, History of Philosophy, Vol. 1

214 Hellenistic-Roman Thought. [Part IL

presentation of their dogmas, led ultimately to the doctrine of Philo

of Alexandria, whose original elaboration of these fermenting bodies

of thought influenced their further formation and movement in the

most important points.

The philosophy of Christianity, which for these first centuries is

usually designated by the name Patristics, unfolded in an analogous

manner upon a larger scale. This philosophical secularisation of the

gospel begins with the Apologists, who sought to present its re-

ligious belief as the only true philosophy, with the purpose of pro-

tecting Christianity in the eyes of the cultured world from contempt

and persecution, and therefore began to adapt this content of re-

ligious faith to the conceptional forms of Greek science : the most

important of them are Justin and Minucius Felix.

But the need of changing faith (71-10-1-15) into knowledge or wisdom

(yvwo-is) asserted itself vigorously in the Christian communities,

even without this polemical tendency. The first attempts, how-

ever, which the Gnostics made to create an adequate view of the

world for the new religion, proceeded from the excited phantasies

of a Syrian mingling of religions, and, in spite of the employment of

Hellenistic philosophemes, led to such grotesque constructions, that

the Church as it grew stronger and more definitive was obliged to

reject them. Satuminus, Basileides, and Valentinus are to be namedas the best known of this class.

In reaction against such over-hasty attempts of religious fantasti-

calness, a violent aversion toward all philosophical interpretation

and adjustment of Christian faith set in, for a time, in Christian

literature in the writings of men like Tatian, Tertullian, and Arno-

bius. An express anti-rationalism thus came forward which never-

theless found it necessary on its part also to return to the related

doctrines of Greek philosophy. Without this one-sidedness and

with a closer approximation to the older Hellenising Apologists,

Gnosticism was combated by Irenazus and his disciple Hippolytus.

It was not until the beginning of the third century, and after all

these preceding attempts, that a positive Christian theology, a sys-

tem of dogmatics in a complete conceptional form, was established.

This came about in the School for Catechists at Alexandria, through

the leaders of the school, Clement and Origen. The latter especially

is to be regarded as philosophically the most important representa-

tive of Christianity in this period.

By his side, however, there went out from the Alexandrian phil-

osophic school the man who undertook to bring the religion-forming

tendency of philosophy to an issue solely upon the Hellenistic basis,

— Plotinus, the greatest thinker of this period. His attempt to

Page 235: Windelband, History of Philosophy, Vol. 1

Chai'. 2.] The Retigious Period. 215

systematise all the main doctrines of Greek and Hellenistic phil-

osophy under the religious principle is designated as Neo- Platonism.

His doctrine is the most definitive and thoroughly constructed sys-

tem of science that antiquity produced. His disciple Porphyry,

however, showed himself already inclined to make a religion out of

this religious teaching, and Jamblichus, who is termed the leader of

Syrian Neo-Platonism, transformed it into a dogmatic theology ofpoly-

theism, with which the learned and political opponents of Christianity,

such as the Emperor Julian, hoped to revive the forms of worship

of the heathen religions, then in a state of dissolution. After this

attempt had miscarried, the Athenian school of Neo-Platonism, as

the heads of which Plutarch of Athens, Proclus, and Damascius

appear, returned finally to a methodical, scholastic development of

the system of Plotinus.

Thus the Hellenistic efforts to attain to a new religion by meansof science remained without result in this form : the scholars dis-

covered no church. On the other hand, the need felt by positive

religion to complete and strengthen itself in a scientific doctrine did

attain its goal : the Church created its dogma. And the great course

of history in this movement was, that the defeated Hellenism in its

powerful death-struggle still created the conceptions by means of

which the new religion shaped itself into a dogma.

While the Pythagorean mysteries had maintained their existence through all

antiquity, scientific Pythagoreanism vanished as a proper school after its

incorporation into the Academy (cf. p. 31). It is not until during the first

century b.c. that specifically Pythagorean doctrines become noticeable again :

they appear in the Pythagorean writings, of which Diogenes Laertius (VIII.24 If.), following Alexander Polyhistor, gives an account that leads us to infer

an essentially Stoic influence. They are renewed expressly by Cicero's learnedfriend, P. Nigidius Figulus (died 45 b.c.), and find approval also with othermen in Rome. Cf . M. Herz, De P. Nig. Fig. Studiis atque Operibus (Berlin, 1845).

But Neo-Pythagoreanism proper was first presented in literary form bythe great number of writings which became public in Alexandria at about thebeginning of our era, under the names of Pythagoras, or Philolaus, or Arcliytas,

or other older Pythagoreans, the fragments of which give rise to so great diffi-

culties in forming a conception of genuine Pythagoreanism. Cf. the lit. p. 31.

Of the personalities of the new school, on the contrary, very little is known.The only distinct figure is Apollonius of Tyana, of whose life and nature therhetorician Philostratus (ed. by C. L. Kayser, Leips. 1870) gave a romanticrepresentation at the beginning of the third century, in order to portray in it

the ideal of the Pythagorean life. Of the works of Apollonius himself, wholived in the first century a.D., fragments of a biography of Pythagoras and ofa treatise on Sacrifice are extant. Cf. Chr. Baur, Apollonius und Christus inDrei Abhandl. zur Gesch. d. alt. Philos. (Leips. 1876). [Tredwell, Life ofApollonius of Tyana, contains a good bibliography, N.Y. 188(5. J His con-temporary, Moderatus of Gades, might perhaps also be mentioned.

Neo-Pythagorean and Stoic doctrines appear mingled in the Eclectic Sotionof Alexandria, who was affiliated with the Sextians (cf. p. 168). His disciple,L. Annaeus Seneca of Cordova (4-65 a.D.), was the leader of the Stoics inthe time of the Empire. He was instructor of Nero, was well known because othis tragic fate, and also as tragic poet unfolded the rigid conceptions of life held

Page 236: Windelband, History of Philosophy, Vol. 1

216 Hellenistic-Roman Thought. [Part II.

by his school. Of his writings a considerable number of mainly ethical trea-

tises are preserved besides his Epistolce (ed. by Haase, 3 vols., Leips. 1852-3)[Eng. tr. (or rather paraphrase) by T. Lodge, Lond. 1614, Selections from this

and from L'Estrange's Seneca's Morals by Way of Abstract, Lond. 1888, Came-lot series]. Cf. Chr. Baur, S. und Paulus in the Drei Abhandl.; see above.

Besides him we mention L. Annseus Cornutus (Phurnutus), a chief repre-

sentative of the Stoic interpretation of myths (Ilept tt)s tCov deCov (pvcrews, ed.

by Osann, Göttingen, 1844), the satiric poet Persius, the moralist C. MusoniusRufus, and especially Epictetus, who lived at the time of Domitian, and whosedoctrines were published by Arrian in two works, Aiarpißat and 'Eyxe'-P^iov (ed.

together with the commentary of Simplicius by J. Schweighauser, Leips. 1799 f.)

[tr. by G. Long, Bonn's library ; also by T. W. Higginson, Boston, 1865]. Cf.

A. Bonhöffer E. und die Stoa (Stuttgart, 1890).

With the noble Marcus Aurelius Antoninus the Stoa mounted the Romanimperial throne (161-180). His reflections rd et's avrbv (ed. by J. Stich, Leips.

1882) are the characteristic monument of this eclectic-religious Stoicism.

[Eng. tr. by G. Long. The Thoughts of the Emperor, M. Aurelius Antoninus,Lond. Bonn's lib. ; W. Pater, Marius the Epicurean, Lond. and N.Y. 1888 ; M.Arnold in Essays.]

In the ancient Grecian period, an original figure, that of the monkish wan-dering preacher Teles, had gone out from the Cynic school (cf. v. Wilamovitz-Möllendorf, Philol. Unters, IV. 292 ff.). In the time of the Empire this quaintcreature was frequently copied and exaggerated even to the most ridiculous

extent. Demetrius, Oinomaos of Gadara, Demonax (cf. Fritsche, Leips. 1866),

and Peregrinus Proteus, known through Lucian, belong to these figures. Cf.

J. Bernays, Lukian und die Kyniker (Berlin, 1879).

Of the representatives of religious Platonism who kept at a distance fromthe number theory, may be mentioned the eclectic commentators Budorus andArius Didymus, Thrasyllus, the editor of the works of Plato and Democritus,and especially Plutarch of Chaeronea (about 100 a.d.), from whom, in addition

to his famous biographies, a great number of other writings are preserved,

especially philosophical treatises of dogmatic and polemical content {Moralia,ed. Diibner ; Paris, Didot, Vols. III. and IV. 1855) (cf. R. Volkmann, Leben,

Schriften und Philosophie des P., Berlin, 1872). [Plutarch's Morals, trans, ed.

by Goodwin, 5 vols., Boston, 1870 ; also tr. by Shilleto and by C. W. King, bothin Bohn's lib., Lond. 1888 and 1882 resp.] We mention further Maximus of

Tyre of the time of the Antonines ; his contemporary, Apuleius of Madaura,who belongs in this series not only on account of his philosophical writings (ed.

by A. Goldbacher, Vienna, 1876), but also on account of his allegorico-satirical

romance, "The Golden Ass" (cf. Hildebrand in the introduction to his col-

lected works, Leips. 1842) [The Works of Apuleius, Bohn's lib.]; the oppo-nent of Christianity, Celsus, whose treatise aXr/dys \6yos (about 180) is knownonly from the counter-treatise of Origen, /card K4\<rov (cf. Th. Keim, C. " wahresWort,'''

1 Zürich, 1873); and lastly the physician Claudius Galen, who died about

200, and might, to be sure, with his broad eclecticism be likewise classed as a Peri-

patetic and also as a Stoic (cf. K. Sprengel, Beiträge zur Gesch. d. Mediän, I.

117 ff.). From the same circle of ideas arose also the writings circulated underthe name of Hermes Trismegistus, which belong to the third century (Frenchtr. by L. Me'nard, Paris, 1866

;partially published by G. Parthey, Berlin, 1854).

Among the Platonists of the second century Nicomachus of Gerasa in Ara-

bia, of whose writings arithmetical text-books and (through Photius) an extract

from a work ' Apid^TiKä deoXoyovßeva are extant, and Numenius of Apamea,concerning whom we owe our instruction mainly to Eusebius, are strongly Neo-Pythagorean. Cf. F. Thedinga (Bonn, 1875).

The entrance of Greek philosophy into Jewish theology may be traced backto the middle of the second century b.c., where it can be recognised in the

Biblical explanation of Aristobulus ; it appears then in a particularly markedmanner, and in a form that is already much nearer the Alexandrian sphere of

thought, in the pseudo-Solomonic Book of Wisdom. Yet these are but weakforerunners of the important creation of Philo of Alexandria, of whose life

little more is known than that in the year 39, when already in advanced age, he

was a member of an embassy from his native community to the Emperor Calijj

Page 237: Windelband, History of Philosophy, Vol. 1

Chap. 2.] The Religious Period. 217

ula. His numerous writings, among which there is also much that is not

genuine, were edited byTh. Mangey (Loud. 1742), Leips. stereotype ed., 8 vols.,

1851-53;[Eng. tr. by C. 1). Yonge, 4 vols., Loud., Bonn's lib.].

F. Dähne, Die yüdisch-alexandrinische Religionsphilosophie (Halle, 1834).

A.Gfrörer, Philon und die alexandrinische Theosophie (Stuttgart, 1835); M.Wolff, Die philonische Philosophie (Gothenburg, 1858); Ewald, Gesch. des

Volkes Israel, VI. 231 ff.

Among the Christian Apologists whose writings are collected in the CorpusApoloyetarum Christianorum secundi sccculi, ed. by Otto (Jena, 1842 ft".), the

most prominent is Elavius Justin Martyr of Sichern, who lived in the middle of

the second century. Two defensive writings and a dialogue with Trypho the

Jew are preserved [Eng. tr. in Ante-Nicene Ch. lib., ed. by Roberts and Donald-son, Edinburg, T. & T. Clark, 1867—]. K. Semisch (2 vols., Breslau, 1840-42),

and B. Aub6 (Paris, 18(51) treat of him. Further Apologists from the Hellenic

circle of culture are Aristides (whose discourses, discovered in the Armenianlanguage, were printed with a Latin translation, Venice, 1878), Athenagorasof Athens (irpea-ßela irepl Xp«rTt.ai>Qv addressed to Marcus Aurelius about 176),

Theophilus of Antioch (a treatise addressed to Autolycus about 180), Melitoof Sardis, Apollinaris of Hierapolis, and others.— Latin literature presents

especially Minucius Felix, whose dialogue Octavitis was written about 200(ed. in the Corpus scriptorum ecclesiasticorum latinorum, by C. Halm, Vienna,

1867). The rhetorician, Firmianus Lactantius (about 300), is to be placed in

the same series. His main treatise is the Institutiones Divince [tr. of the aboveauthors in Ante-Nicene lib., see above].

Of the Gnostics our information comes essentially through their opponents,

Irena?us (140-200 ; his treatise "EXe7xos Kai ävarpowr] rijs \l/evdojvvp.ov yvwcrews, ed.

by A. Stieren, Leips. 1853), Hippolytus (Kara iracrwv aipfoewv ZXeyxos, ed. byDuncker and Schneidewin, Göttingen, 1859), Tertullian (Adversus Valenti-

nianos), etc. [Eng. tr. of the above writings in Ante-Nicene lib., above]. OfGnostic treatises only one, and that by an unknown author, is extant, Marisao<pla (ed. by Petermann, Berlin, 1851). Of the main representatives of this

doctrine there were active in the first half of the second century Saturninus of

Antioch, Basilides, a Syrian, and Carpocrates in Alexandria ; toward the

middle of the century Valentinus, the most important of them (died about

160); and toward the end of the century Bardesanes of Mesopotamia.— Expo-sitions of the Gnostic Systems by A. W. Neander (Berlin, 1818) [Eng. tr. byTorrey, Boston, 1865], E. Matter (Paris, 1843), Chr. Baur (Tübingen, 1835),

A. Hilgenfeld (Jena, 1884), same author, Bardesanes, der letzte Gnostiker(Leips. 1864). — A. Harnack, Zur Quellenkritik der Geschichte des Gnosticismus(Leips. 1873); [H. L. Mansel, Gnostic Heresies, Lond. 1876].

The most radical opponent of Greek science was Tatian, an Assyrian,

whose treatise Upbs''EXXtjws arose about 170, but who later became himself anadherent of the Valentinian Gnosticism. The passionate Apologist Qu. Sep-

timius Florens Tertullian (165-220, for a time Presbyter in Carthage) endedlikewise in opposition to the Catholic Church, in the sect of the Montanists.

His works have been edited by Fr. Oehler (3 vols., Leips. 1853 f.), recently byA. Reifferscheid and Wissowa (Vol. I. Vienna, 1890, in Corp. script, eccl. lat.)

[Eng. tr. in Ante-Nicene lib.]. Cf. A. W. Neander, Antignosticus, Geist des

Tertullian, etc. (2d ed. Berlin, 1849) [Eng. tr. Bonn's lib., 1851]; A. Hauck,T.'s Leben und Schriften, Erlangen, 1877).

In the same series, but froma later time, is the African rhetorician Arnobius, whose seven books, AdversusGentes, were composed about 300 (ed. by A. Reifferscheid in Corp. script, eccl.

lat., Vienna, 1875).

Of the writings of Clement of Alexandria (died about 217) three treatises

are preserved, A670S TrporpeirTiicds irphs "EWwvas— YlaidaywySs— ^,Tpwfxare7s (ed.

by J. Potter, Oxford, 1715) [tr. in Ante-Nicene lib.]. From his school (cf. onthe Alex. Catechetical school, Guericke, Halle, 1824 f., and Hasselbach, Stettin,

1820) went forth the founder of Christian theology, Origen, sumamed the Ada-mantine. Born 185 a.d. in Alexandria, equipped with the full education of the

time, he came forward early as a teacher, fell into conflicts on account of his

doctrines with the Synod, was by it removed from his office, and later lived in

Caesarea and Tyre, dying in the latter place 254. Of his writings, aside fromthe above-mentioned treatise against Celsus, his work Ilepl äp\Qv is of chief

importance ; it is extant almost only in the Latin version of Rutinus (ed. by

Page 238: Windelband, History of Philosophy, Vol. 1

218 Hellenistic-Roman Thought : Religious Period. [Part II

Redepenning, Leips. 1836) [tr. in Ante-Nicene lib.]. Cf. J. Reinkens, DeClemente Presbytero AI. (Breslau, 1851); Redepenning, 0., Darstellung seinesLebens und seiner Lehre (Bonn, 1841-4(5) [cf. Bigg, The Christian Platonists

of Alexandria, Oxford, 1887 ; A. Harnack, Art. Origen in Enc. Brit.].A collection of the sources for all the Church writers of this period has been

issued by J. P. Migne, Patrologice Cursus Computus (Paris, 1840 ff.).

A certain Ammonius Saccus appears in old traditions as the founder ofNeo-Platonism, but nothing is known to justify this tradition. To his pupilsbelonged Plotinus, Origen, the rhetorician Longinus (213-273), to whom thebook Ilepl tyovs was ascribed, and another Origen.

The true founder of the school was Plotinus (204-269). Born in Lycopolisin Egypt, and educated in Alexandria, he became a member of an expeditionagainst the Persians in order to promote his religious studies, made a highlysuccessful appearance as teacher in Rome about 244, and died on a countryestate in Campania. His works, written late in life, were published by his

disciple Porphyry, arranged in six enneads. Ed. by H. Müller (Leips. 1878-80),with a German translation [Eng. tr. in part by Th. Taylor, Lond. 1787, 1794,

1817, French tr. by Bouillet, Paris, 1857-60]. Cf. H. Kirchner, Die Philos. des

PI. (Halle, 1854).— A. Richter, Neuplatonische Studien (Halle, 1864 ff.).—H. v. Kleist, Neuplat. Studien (Heidelberg, 1883). — [A. Harnack, Art. Neo-Platonism in Enc. Brit.]

To the Alexandrian Neo-Platonism are reckoned further Gentilianus Ame-lius of Ameria, and the Tyrian Porphyry (about 230-300). Among the ex-

tant writings, aside from the biographies of Plotinus and Pythagoras, are to bementioned 'Acpoppal rrpbs rä vo-qrä, an aphoristic abridgment of the system of

Plotinus (printed in Creuzer's ed. of the works of Plotinus, Paris, 1855), the

treatise On Abstemiousness (irepl äwoxv* T&v iß^Cxuv> important on account of

its use of the irepl evaeßelas of Theophrastus ; cf. J. Bernays, Berlin, 1866), andof the commentaries the 'ElcrayuyT) eis ras Karriyoplas (ed. by Busse, Berlin, 1877 •

and also in the Berlin ed. of Aristotle, Vol. IV.).

Syrian Neo-Platonism was founded by Jamblichus of Chalcis in Ccele-

Syria (died about 330), a hearer of Porphyry. His writings were principally

commentaries upon Hellenistic and Oriental theology. The following are par-

tially preserved : Ilepl rod YlvdayopiKoO ßlov (ed. by Westermann, Paris, 1850),

A670S irporpeTTTiKÖs els <pi\o(ro<piav (ed. by Kiessling, Leips. 1813), Ilepl rijs tcoivrjs

/j.a6^fiaTLK7js iTTLaT7jiJ.T]s (ed. by Villoison, Venice, 1781) [Eng. tr. Life of Pyth.

by Taylor, Lond. 1818, Egyptian Mysteries, by same, Chiswick, 1821].

Of the disciples of the school, Dexippus commented on the Aristotelian

Categories (ed. by L. Spengel, Munich, 1859), Sallustius wrote a compendiumof metaphysics (ed. by Orelli, Zürich, 1821), and Themistius (about 317-387)

made himself known as a paraphrast and commentator upon Aristotelian works.

From the same circle comes the treatise De Mysteriis ^Egyptiorum (ed. by G.

Parthey, Berlin, 1857 ; cf. Harless, Munich, 1858).

This movement had a transient political success by the accession of the

Emperor Julian, who hoped by its help to renew the old religion and displace

Christianity. His writings against the Christians have been edited with aGerman translation by K. J. Neumann (Leips. 1880). Cf. A. W. Neander,

TJeber den Kaiser J. und sein Zeitalter (Berlin, 1812). — D. Fr. Strauss, J. der

Abtrünnige, der Bomantiker auf dem Throne der Cäsaren (Mannheim, 1847).

A. Mücke, J. nach den Quellen (Gotha, 1866-68).

The founder of Athenian Neo-Platonism was Plutarch of Athens (died

after 430), with his pupils Syrianus and Hierocles. All these, as well as the

following, composed commentaries upon Platonic and Aristotelian or Pythago-

rean writings, which are in part preserved. More important was Proclus(411-485), among whose works the most important is Ilepl rijs Kara HXdrwvajeoXoyias (ed. of his works by V. Cousin, Paris, 1820-25) [Eng. tr. by Th.

Taylor]. Cf. H. Kirchner, De Prod. Metaphysica (Berlin, 1846). K. Stein-

hart's Art. in Erschund Grüber's Enc.The last head of the Platonic Academy was Damascius, of whose writings

the beginning of a treatise irepl tGjv -Kpibrwv &px&v, and the conclusion of a com-mentary upon the Parmenides are extant (ed. by J. Kopp, Frankfort a. M.182a ; cf. E. Heitz in Strass. Abhdl. für Philos., 1884), and also a biography of

Page 239: Windelband, History of Philosophy, Vol. 1

c'hap. 2, § 18.] Authority and Revelation. SJ19

his teacher Isidorus. Among the commentators of this time Simplicius is

prominent (on the Physics, ed. pr. Venice, 1526, the first four books, Diels,

Berlin, 1882 ; on the De Cailo, Karsten, Utrecht, 18(35 ; on the De A nima,Hayduck, Berlin, 1882).

The two latter wandered with their immediate associates for a time towardPersia, when in the year 529 the Emperor Justinian closed the Academy, con-fiscated its property, and by forbidding lectures on heathen philosophy gavethe external confirmation to its close.

§ 18. Authority and Revelation.

The imperturbable self-certainty and self-mastery which the post-

Aristotelian philosophy had sought and in part claimed for the wise

man, had been so deeply shaken with the progress of time that it

had given place to a feeling of the need of help, both in the ethical

and in the theoretical spheres. The philosophising individual no

longer had confidence that he could attain to right insight or to his

soul's salvation by his own strength, and sought his help accord-

ingly, partly amid the great monuments of the past, partly in

a divine revelation. Both tendencies, however, are ultimately upon

the same basis, for the confidence which was placed in the men and

writings of a previous time rested only upon the fact that they

were regarded as especially favoured vessels of higher revelation.

Authority, therefore, acquired its value as the mediate, historically

accredited revelation, while the divine illumination of the individ-

ual as immediate revelation came to its assistance. Differently as

the relation between these two forms was conceived of, it is yet the

common mark of all Alexandrian philosophy that it regards divine

revelation as the highest source of knoivledge. Already in this inno-

vation in the theory of knowledge, we find expressed the heightened

value which this period put upon personality, and on personality as

evincing itself in the feelings. The longing of this time desired

that the truth might be found by experience, as an inner commun-

ion of man with the Supreme Being.

1. The appeal to authority often makes its appearance in Greek

and Hellenistic philosophy in the sense of a confirmation and

strengthening of an author's own views, but not as a decisive and

conclusive argument. The jurare in verba magistri might be usual

enough among the subordinate members of the schools, 1 but the

heads of schools, and in general the men who engaged in indepen-

dent research, maintained an attitude toward the teachings of the

former time that was much more one of criticism than of uncondi-

tional subjection;

2 and though in the schools, chiefly the Academic

1 Though even the well-known avrbs ipa. [ipse dixit] of the Pythagoreans is

attested only through later writers (Cicero).2 Even the admiration of Socrates, in which all the following schools were at

one, did not in itself lead to his being regarded as the valid authority for defi-

nite philosophical doctrines.

Page 240: Windelband, History of Philosophy, Vol. 1

220 Hellenistic-Roman Thought : Religious Period. [Part II.

and Peripatetic, the inclination to preserve and maintain the

teaching of the founder as an unassailable treasure was fostered by

the custom of commenting upon his works, yet in all the conflict

as to the criteria of truth the principle had never been brought

forward that something must be believed because this or that great

man had said it.

How strongly the need for authority had come to be felt in the

later time, we may recognise even from the countless interpolations

which were the order of the day in the whole Alexandrian litera-

ture. Their authors, who, perhaps, for the most part acted in good

faith, since they themselves regarded their thoughts as only devel-

opments and continuations of the old doctrines, evidently believed

that they could get a hearing for their works in no better way than

by assigning to them the name of one of the heroes of wisdom, of

an Aristotle, a Plato, or a Pythagoras. This phenomenon appeared

most extensively among the !N~eo-Pythagoreans, whose chief con-

cern it was to invest their new doctrine with the halo of ancient

wisdom. But the more the convictions that were to be established

in this manner bore a religious character, the more lively became

the need to conceive of these authorities themselves as the bearers

of a religious revelation, and therefore all the traits that might

stamp them as such were sought for within them or even read into

them. Not contented, however, with this, the later Greeks believed

that they could give a higher sanction to their philosophy, as well

as to their entire civilisation, by deriving it from the Oriental

religions : thus Numenius 1 did not hesitate to maintain that

Pythagoras and Plato had presented only the old wisdom of the

Brahmans, Magi, Egyptians, and Jews. As a result of this, the

extent of literary authorities increased extraordinarily ; the later

Neo-Platonists, a Jamblichus and Proclus, commented not only on

Greek philosophers, but also upon the entire Hellenic and barbarian

theology,2 and credulously adopted myths and miraculous tales

from these sources.

In quite a similar manner Oriental literature testified also to its

esteem for Hellenism. Among the predecessors of Philo, Aristo-

bulus especially appealed to verses which were interpolated in

Orpheus and Linus, in Homer and Hesiod ; and with Philo himself,

the great Jewish theologian, the great men of Greek philosophy

appear side by side with the Old Testament, as bearers of wisdom.

The felt need of authority naturally asserts itself most strongly

in the unconditional faith in relisrious records. Here the Old Testa-

1 In Eus. Prazp. Ev. IX. 7 2 Marinus, Prod. Vit. 22.

Page 241: Windelband, History of Philosophy, Vol. 1

Chai\ 2, § 18.] Authority and Revelation : Philo. 221

nient was from the beginning the firm foundation for the science

and philosophy of Judaism and also for that of (orthodox) Christian-

ity. But in the Christian Church the need of establishing a collec-

tion of writings in which the system of faith should be defined withcertainty, first developed with Marcion, and then was gradually

satisfied in the completion and conclusion of the New Testament

:

with Irenaeus and Tertullian both Testaments already appear with

the full value and validity of churchly authority.

2. If now in this way even scientific thought, which in conse-

quence of sceptical disintegration no longer gave itself credit for

the power of truth, subjected itself voluntarily to the authorities of

antiquity and to religious institution, it was yet in nowise boundthereby to the extent that we might suppose. This relation rather

took the form, along all lines, of extracting from the authoritative

sources, and also of reading into them, the scientific doctrines whicharose from the new religious movements. 1

Where in so doing they did not resort expressly to those inter-

polations which are found more or less in the entire literature of

the period as well as in Neo-Pythagoreanism, they employed as

their instrument the method of allegorical interpretation.

This meets us first in Jewish theology. It h;id its prototype

indeed in the allegorical interpretation of myths, which made its

appearance early in Grecian literature, was employed by the Sophists,

and extensively prosecuted by the Stoics. It was applied to relig-

ious documents by Aristobulus, but it was Philo 2 who carried it

through methodically, proceeding from the conviction that a dis-

tinction must be made in Scripture between the literal and the

spiritual meaning, between its body and its soul. In order to teach

his commands to the great mass of men, who in their sensuous

nature are unable to apprehend the divine purely, God gave to

revelation the anthropomorphic form, behind which only the spirit-

ually mature man penetrates to the true sense. This sense is to be

sought in the philosophical conceptions which lie hidden in the

historical husks. Accordingly, since Philo the task of theology

has been directed toward interpreting religious documents into a sys-

tem of scientific doctrines; and if he uses Greek philosophy for this

purpose, and finds in it the higher meaning of the Scripture, he

1 Even a man like Plutarch of Chaeronea, who follows the writings of Platoas he would the revelations of a religious document, does not scruple to intro-

duce into the teaching of his master Aristotelian and Stoic doctrines as well ashis own religious view.

2 Cf. Siegfried, Philon v. Alexandria als Ausleger des alten TestamentsrJena, 1875).

Page 242: Windelband, History of Philosophy, Vol. 1

222 Hellenistic-Roman Thought : Religious Period. [Part II.

explains this relation on the ground that the thinkers of Greece

have drawn from Mosaic documents. 1

Following his example, the Gnostics then attempted to transform

Oriental myths into Greek conceptions by allegorical interpretation,

and thought thus to develop a secret doctrine of the Apostolic

tradition,— the Apologists maintained the harmony of Christian

doctrine with the dogmas of Greek philosophy, — even men like

Irenseus and Tertullian worked upon the New Testament,— and

finally Origen knew how to bring the philosophy of Christianity

into accord with its documents. The great Alexandrian theologian,

like the Gnostics who first attempted to create a Christian theology,

distinguished between the carnal (somatic), psychical, and spiritual

(pneumatic) conceptions of the religious records,— corresponding

to the metaphysico-anthropological ideas of the time (cf. § 19 f.).

For him the literal historical tradition yields only a " Christianity

according to the flesh " (^pio-navicr/nos o-w/jlcltlkos) , and it is the task of

theology to lead out of this, through the moral significance at which

the " psychical " readers stop, to the ideal content of the Scripture,

which must then illumine the reader as self-evident truth. Only he

who grasps this last belongs to the pneumatic or spiritual readers,

to whom the eternal gospel thus disclosed reveals itself.

This extraction of philosophical meaning from religious tradition

is found in fullest extent among the Neo-Platonists. Jamblichus

practises it, in accordance with the Stoic model, on all forms of

Oriental and Occidental mythology, and Proclus, too, declares ex-

pressly that myths veil the truth from sensuous men who are not

worthy of it.2

3. But in all such doctrines, the interest of science (in the Chris-

tian teachings, yvwo-i?) ultimately predominates over that of faith

;

they are accommodations of philosophy to the need of religious

authority, felt at this time. The essential identity of authority and

ofrational knowledge obtains, therefore, as the fundamental presuppo-

sition ; it obtains in such a degree, that just where it seems threat-

ened, all artifices of allegorical interpretation are attempted in order

to rescue it. This confidence, nevertheless, with which science pro-

ceeded to develop its own content as that of the religious documents,

rested ultimately upon the conviction that both historical authority

and scientific doctrine are but different revelations of the same divine

Power.

We have seen that the belief in authority in this period grew out

of the felt need of salvation and help. Another psychological root of

i Phil. Vit. Mos. 657 a. (137 m.). a procl. In Bemp. 369.

Page 243: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 18.] Authority and Revelation: Origen, Justin. 22*

this belief was the enhanced importance of personality. This show:-

itself in the lively expression of admiration for the great men of the

past, as we find it in Philo and in all lines of Platonism, and not

less in the unconditional trust of the disciples in their masters,

which, especially in later Neo-Platonism, degenerated to exaggerated

veneration of the heads of schools. 1 This same motive appears in

grandest form as a power in the world's history, in the stupendous,

overpowering impression of the personality of Jesus. Faith in himwas the uniting bond which held together victoriously the various andmanifold tendencies of early Christianity.

But this psychological motive justified itself to theory by the

consideration that the admired personality was regarded, in teach-

ing and life, as a revelation of the divine World-reason. The meta-

physical and epistemological bases for this were given in Platonism

and especially in Stoicism. Attachment to the Platonic doctrine

that knowledge is recollection, with the turn already expressed in

Cicero that right knowledge is implanted by God in the soul, is innate

within it, the carrying out of the Stoic logos doctrine, and of the

idea contained in it that the rational part of the soul is a consub-

stantial emanation from thb divine World-reason,— all this led to

regarding every form of right knowledge as a kind of divine revela-

tion in man. 2 All knowledge is, as Numenius said,3 the kindling of

the small light from the great light which illumines the world.

It was from this point of view that Justin, especially, conceived

of the relationship maintained by him between the old philosophy

and Christianity, and at the same time conceived the superiority of

the latter. God has indeed revealed himself internally through the

rational nature 4(o-n-epfjua \6yov Z/xcfrvrov) of man who is created in

his image, as he has revealed himself externally through the perfec-

tion of his creation ; but the development of this universal, more

potential than actual revelation, is retarded by evil demons and

man's sensuous impulses. God has, therefore, for man's help, em-

ployed the special revelation, which has appeared not only in Moses

and the prophets, but also in the men of Greek science.5 Justin

calls the revelation which is extended to the entire human race, the

1 From the point of view of the history of civilisation we may notice the

parallel in the boundless deification of the Roman Emperors.2 So also by the Stoics of the time of the Empire, philosophy, which among

them likewise aimed to be a cure for sick souls (Epictetus, Dissert. III. 23, 30),

is set forth as a sermon of the deity himself, through the mouth of the wiseman (ib. I. 36).

3 In Euseb. Prcep. Ev. XI. 18, 8.

* Apol. II. 8 ; cf. Min. Fei. Oct. 10, 5.6 On the other hand, to be sure, Justin as well as Thilo derives the Greek

ahilosophy from the Jewish religion, as a borrowing.

Page 244: Windelband, History of Philosophy, Vol. 1

224 Hellenistic-Roman Thought : Religious Period. [Part II.

\oyos o-irepfiaTiKos. But that which has appeared in former time, so

dispersed and often obscured, is not the full truth : the entire, pure

logos has been revealed in Christ, Son of God, and second God.

In this teaching there prevails, on the one hand, with the Apolo-

gists, the effort to set forth Christianity as the true and highest phil-

osophy, and to show that it unites in itself all teachings x of abiding

worth that can be discovered in the earlier philosophy. Christ is

called the teacher (SiSao-KaAos), and this teacher is Reason itself.

While Christianity was by this means brought as near as possible to

rational philosophy, and philosophy's principle of knowledge made

essentially equivalent to that of religion, this had yet at the same

time the consequence, that the conception of the religious content

itself became strongly rationalistic with Justin and similar Apolo-

gists, such as Minucius Felix: the specifically religious elements

appear more repressed, and Christianity takes on the character of a

moralising deism, in which it acquires the greatest similarity to

religious Stoicism. 2

On the other hand, in this relation the self-consciousness of

Christianity speaks out, for with its perfect revelation it regarded

all other kinds of revelation, universal as well as particular, as super-

fluous ; and at this point the Apologetic doctrine became of itself

polemic, as is shown especially in Athenagoras. Revelation here,

too, is still regarded as the truly reasonable, but just on this account

the reasonable is not to be demonstrated, but only believed. Phil-

osophers have not found the full truth, because they have not been

willing or able to learn God from God himself.

4. Thus, although in the Apologetic doctrine the rational is re-

garded as supernaturally revealed, there is gradually preparing an

opiposition between revelation and knoioledge by the reason. The more

the Gnostics, in developing their theological metaphysics, separated

themselves from the simple content of Christian faith, the more

Irenaius 3 warned against the speculations of worldly wisdom, and

the more violently Tatian, with Oriental contempt of the Greeks,

rejected every delusion of the Hellenic philosophy which was

always at variance with itself, and of whose teachers each would

exalt only his own opinions to the rank of law, while the Christians

uniformly subjected themselves to the divine revelation.

This opposition becomes still sharper with Tertullian and Arno-

bius. The former, as Tatian had already done in part, adopted the

1 Apol. II. 13, Sera irapa. iracri /caXtDs etprjTai rjfiQv "KpicmavGiv icrriv.

2 Cf. Min. Fei. Oct. 31 ff., where the Christian fellowship of love appears pre-

cisely as the Stoic world-state of philosophers.3 Ref. II. 25 ff.

Page 245: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 18.] Authority and Revelation : Tertullian, Plutar^a. 22ö

Stoic materialism in its metaphysical aspect, but drew from it only

the logical consequence of a purely sensualistic theory of knowledge.

This was carried out in an interesting way by Arnobius, when, to

combat the Platonic and Platonising theory of knowledge, he showed

that a man left in complete isolation from his birth on would re-

main mentally empty, and not gain higher knowledge. 1 Since the

human soul is by nature limited solely to the impressions of the

senses, it is therefore of its own power absolutely incapable of

acquiring knowledge of the deity, or of any vocation or destiny of

its own that transcends this life. Just for this reason it needs rev-

elation, and finds its salvation only in faith in this. So sensualism

here shows itself for the first time as basis for orthodoxy. The lower

the natural knowing faculty of man, and the more it is limited to

the senses, the more necessary does revelation appear.

Accordingly, with Tertullian, the content of revelation is not only

above reason, but also in a certain sense contrary to reason, in so

far as by reason man's natural knowing activity is to be understood.

The gospel is not only incomprehensible, but is also in necessary

contradiction with worldly discernment : credibile est quia inep-

tum est ; certum est, quia impossibile est— credo quia absurdum. Hence

Christianity, according to his view, has nothing to do with philoso-

phy, Jerusalem nothing to do with Athens. 2 Philosophy as natural

knowledge is unbelief ; there is therefore no Christian philosophy.

5. Put rationalistic theory also found occasions enough for such

a defining of boundaries between revelation and natural knoivledge.

For by their identification the criterion of truth threatened to

become lost. The quantity of that which presented itself as reve-

lation, in this time of such agitation in religion, made it indispen-

sable to decide on the right revelation, and the criterion for this

could not be sought in turn in the individual's rational knowledge,

because the principle of revelation would be thereby injured. This

difficulty made itself very noticeable, especially in the Hellenistic

line of thought. Plutarch, for example, who regards all knowledge

as revelation, follows the Stoic division of theology into three kinds,

— viz. of the poets, of the law-givers, and of philosophers, — and

would concede to science or philosophy the supreme decision as to

religious truth,3 declaring himself vigorously against superstition 4

1 Arn. Adv. Gent. II. 20 ff.

2 Tertull. De Came Chr. 5 ; De Pncscr. 7. In the latter passage lie directs

his polemic also expressly against those who present a Stoic or Platonic Chris-

tianity. He is the extreme opponent of the Hellenising of dogma ; he knowsno compromise, and with his hot-blooded nature demands unconditional surren-

der to revelation. In a still more popular manner Arnobius sets forth the help-

lessnpss of natural knowledge (Adv. Gent. II. 74 ff.).

3 Dp Isid. 08. * De Superst. 14.

Page 246: Windelband, History of Philosophy, Vol. 1

226 Hellenistic-Roman Thought : Religious Period. [Part II

(SeicnScu/Aovta) ; but he shows himself to be ultimately as naive and

credulous as his time, since he takes up into his writings all kinds

of tales of prophecies and miracles ; and the incredible absence of

criticism with which the later Neo-Platonists, a Jamblichus and

Proclus proceeded in this respect, shows itself as the consistent

result of the renunciation of the thinker's own discernment, — a

renunciation which the need of revelation brought with it from the

beginning.

Here the development of the Church, which was then in process

of organisation, set in with its principle of tradition and historically

accredited authority. It regards the religious documents of the Old

and New Testaments as entirely, and also as alone, inspired. It

assumes that the authors, in recording this highest truth, were

always in a state of pure receptivity in their relation to the divine

spirit,1 and finds the verification of this divine origin, not in the

agreement of this truth with the knowledge derived from humanreason, but essentially in the fulfilment of the prophecies which are

therein contained, and in the purposeful connection of their succession

in time.

The prooffrom prophecy, which became so extraordinarily impor-

tant for the further development of theology, arose accordingly from

the need of finding a criterion for distinguishing true and false

revelation. Since man is denied knowledge of the future through

natural processes of cognition, the fulfilled predictions of the proph-

ets serve as marks of the inspiration, by means of which they have

propounded their doctrines.

To this argument a second is now added. According to the doc-

trine of the Church, which on this point was supported chiefly by

Irenaeus,2 Old and New Testaments stand in the following connec-

tion : the same one God has revealed himself in the course of time

to man in a constantly higher and purer manner, corresponding to

the degree of man's receptive capacity : to the entire race he

reveals himself in the rational nature, which, to be sure, may be mis-

used ; to the people of Israel, in the strict law of Moses ; to entire

humanity again, in the law of love and freedom which Jesus an-

nounced.3 In this connected succession of prophets there is thus

developed the divine plan of education, according to which the reve-

lations of the Old Testament are to be regarded as preparations for

1 Just. Apol. I. 31.2 Bef. III. 12 ; IV. 11 ff.

3 The Alexandrian theology added, as fourth phase of revelation, the " eter-

nal gospel." which is to be sought in the pneumatic interpretation of the NewTestament. Cf. the carrying out of these thoughts in Lessing's Education ofthe Human Bace.

Page 247: Windelband, History of Philosophy, Vol. 1

Chap. 2, §18.] Authority and Revelation : Neo-Platoniam, Philo. 221

the New, which in turn confirms them. Here, too, in patristic

literature, the fulfilment of prophecies is regarded as the connect-

ing link between the different phases of revelation.

These are the forms of thought in which the divine revelation

became fixed for the Christian Church as historical authority. Butthe fundamental psychological power which was active in this pro-

cess remained, nevertheless, devotion in faith to the person of

Jesus, who, as the sum total of divine revelation, formed the centre

of Christian life.

6. The development of the doctrine of revelation in the Hellenistic

philosophy took an entirely different direction. Here the scientific

movement lacked the living connection with the Church community,

and therefore the support of a historical authority ; here, therefore,

revelation, which was demanded as a supplement for the natural

faculties of knowledge, must be sought in an immediate illumination

of the individual by the deity. On this account revelation is here

held to be a supra-rational apprehension of divine truth, an appre-

hension which the individual man comes to possess in immediate con-

tact (d<t>t]) with the deity itself: and though it must be admitted

that there are but few who attain to this, and that even these attain

only in rare moments, a definite, historically authenticated, special

revelation, authoritative for all, is nevertheless here put aside.

This conception of revelation was later called the mystic conception,

and to this extent Neo-Platonism is the source of all later mysticism.

The origins of this conception again are to be sought with Philo.

For he had already taught that all man's virtue can arise and con-

tinue only through the working of the divine Logos within us, and

that the knowledge of God consists only in the renunciation of self,

— in giving up individuality, and in becoming merged in the divine

Primordial Being. 1 Knowledge of the Supreme Being is unity of

life with him, — immediate contact. The mind that wishes to behold

God must itself become God. 2 In this state the soul's relation is

entirely passive and receptive

;

3it has to renounce all self-activity,

all its own thought, and all reflection upon itself. Even the vous,

the reason, must be silent in order that the blessedness of the per-

ception of God may come upon man. In this state of ecstasy

(«Ktrracrts) the divine spirit, according to Philo, dwells in man.

Hence, in this state, he is a prophet of divine wisdom, a foreteller

and miracle-worker. As the Stoa had already traced mantic arts

i Phil. Leg. All. 48 e. ; 55 d.j 57 b. (53-62 M.).2 'Airo8eiodrji>ai is found also in the Hermetic writings; Poemand. 10, 5 ff.

The deovadai (deificatio) is later a general term of Mysticism.a Cf. Plut. DePyth. Orac. 21 ff. (401 ff.).

Page 248: Windelband, History of Philosophy, Vol. 1

228 Helh'-nistic-Roman Thought : Religious Period. [Part II.

to the consubstantiality of human and divine spirits (nvcvfiaTa) , so

too the Alexandrians conceive of this "deification" of man from

the standpoint of his oneness in essence with the ground of the

world. All thought, Plotinus teaches, is inferior to this state of

ecstasy ; for thought is motion,— a desiring to know. Ecstasy,

however, is certainty of God, blessed rest in him ; * man has share

in the divine Oewpia, or contemplation (Aristotle) only when he has

raised himself entirely to the deity.

Ecstasy is then a state which transcends the self-consciousness of

the individual, as its object transcends all particular determinate-

ness (cf. § 20, 2). It is a sinking into the divine essence with an

entire loss of self-consciousness : it is a possession of the deity, a

unity of life Avith him, which mocks at all description, all percep-

tion, and all that abstract thought can frame.2

How is this state to be attained ? It is, in all cases, a gift of the

deity, a boon of the Infinite, which takes up the finite into itself.

But man, with his free will, has to make himself worthy of this

deification. He is to put off all his sensuous nature and all will

of his own ; he is to turn back from the multitude of individual

relations to his pure, simple, essential nature (a7rA(ocns);

3 the ways

to this are, according to Proclus, love, truth, and faith ; but it is

only in the last, which transcends all reason, that the soul finds its

complete unification with God, and the peace of blessed rapture. 4 As

the most effective aid in the preparation for this operation of divine

grace, prayer 5 and all acts 6 of religious worship are commended.

And if these do not always lead to the highest revelations of the

deity, they yet secure at least, as Apuleius 7 had before this sup-

posed, the comforting and helpful revelations of lower gods and

demons, of saints and guardian spirits. So, also, in later Neo-

Platonisra, the raptures of prophecy which the Stoics had taught

appear as lower and preparatory forms for the supreme ecstasy of

deification. For, ultimately, all forms of worship are to the Neo-

Platonist but exercises symbolic of that immediate union of the

individual with God.

Thus the theory of inspiration diverged, in Christianity and Neo-

Platonism, into two wholly different forms. In the former, divine

1 Plot. Ennead. VI. 7.2 lb. V. 3.8 An expression which is found even with Marcus Aurelius (Ilpbs iavr. IV.

26), and which Plotinus also employs (Enn. VI. 7, 35).4 Procl. Theol. Plat. I. 24 f.

5 Jambl. in Procl. Tim. 64 C.G De Myst. JEg. II. 11 (96)." Apul. De Socr. 6 ff.

Page 249: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 19.J Spirit and Matter: Stoics, Neo-Pythagoreans. 229

revelation is fixed as historical authority ; in the latter, it is the

process in which the individual man, freed from all eternal relation,

sinks into the divine original Ground. The former is for the Middle

Ages the source of Scholasticism ; the latter, that of Mysticism.

§ 19. Spirit ' and Matter.

Among the arguments in which the felt need of revelation devel-

ops in the Alexandrian philosophy, none is so incisive as that which

proceeds from the premise that man, ensnared in the world of sense,

can attain to knowledge of the higher spiritual world only by super-

natural help: in this is shown the religious dualism which forms

the fundamental mode of view of the period. Its roots are partly

anthropological, partly metaphysical : the Stoic antithesis of reason

and what is contrary to reason is united with the Platonic distinction

between the supersensuous world, which remains ever the same,

and the sensuous world which is always changing.

The identification of the spiritual and the immaterial, which was in

nowise made complete with Plato although he prepared the wayfor it, had been limited by Aristotle to the divine self-consciousness.

All the spiritual and mental activities of man, on the contrary, were

regarded, even by Plato, as belonging to the world of phenomena

(yeveo-is), and remained thus excluded from the world of incorporeal

Being (ovaia), however much the rational might be opposed to the

sensuous in the interest of ethics and of the theory of knowledge;

and while, in the antagonistic motives which crossed in the Aristo-

telian doctrine of the vovs, the attempt had been made to regard

Reason as an immaterial principle, entering the animal soul from

without, the development of the Peripatetic School (cf. § 15, 1) at

once set this thought aside again. It was, however, in the doctrines

of Epicurus and the Stoa that the conscious materialising of the

psychical nature and activities attained its strongest expression.

On the other hand, the ethical dualism, which marked off as

strongly as possible, man's inner nature, withdrawn into itself, as

over against the sensuous outer world, became more and more

sharply accentuated, and the more it took on religious form, the

more it pressed, also, toward a theory of the world that made this

opposition its metaphysical principle.

1 [The German " Geist,'1 '' corresponding to both "mind" and "spirit," as

used in this period leans sometimes to one, sometimes to the other meaning.In view of the prevailingly religious character of the ideas of the period I haveusually rendered it in this section by "spirit," sometimes by the alternative" mind or spirit."]

Page 250: Windelband, History of Philosophy, Vol. 1

230 Hellenistic-Roman Thought : Religious Period. [Part U

1. This relation appears in clearest form, perhaps, in the expres-

sions of the later Stoics, who emphasise anthropological dualism so

strongly that it comes into palpable contradiction with the meta-

physics of the school. The idea of the oneness of man's nature,

which the Stoics had taught hitherto, had indeed been already

questioned by Posidonius, when he expressed the Platonising

opinion, that the passions could not arise from the ^ye/xoviKov, but

must come from other irrational parts of the soul. 1 Now, however,

we find in Seneca 2 a bald opposition between soul and "flesh" ; the

body is only a husk, it is a fetter, a prison for the mind. So, too,

Epictetus calls reason and body the two constituent elements of

man,3 and though Marcus Aurelius makes a distinction in man's

sensuous nature between the coarse material and the psychical

breath or pneuma which animates it, it is yet his intention to sep-

arate all the more sharply from the latter the soul proper, the

rational spirit or intelligence (vovs and Stdvoia), as an incorporeal

being. 4 In correspondence with this, we find in all these men an ides

of the deity, that retains only the intellectual marks from the Stoic

conception, and looks upon matter as a principle opposed to the deity,

hostile to reason. 5

These changes in the Stoa are due, perhaps, to the rising influence

of Neo-Pythagoreanism, which at first made the Platonic dualism,

with its motives of ethical and religious values, the centre of its

system. By the adherents of this doctrine the essential difference

of soul and body is emphasised in the strongest manner,6 and with

this are most intimately connected, 7 on the one hand, the doctrine

which will have God worshipped only spiritually, as a purely

spiritual being,8 by prayer and virtuous intention, not by outward

acts,— and on the other hand, the completely ascetic morals which

aims to free the soul from its ensnarement in matter, and lead it

back to its spiritual prime source by washings and purifications, by

avoiding certain foods, especially flesh, by sexual continence, and

by mortifying all sensuous impulses. Over against the deity, which

is the principle of good, matter (rAi?) is regarded as the ground of

all evil, propensity toward it as the peculiar sin of man.

1 Cf. Galen, De Hipp, et Plat. IV. 3 ff.

2 Senec. Epist. 65, 22 ; 92, 13 ; Ad Marc. 24, 5.

3 Epict. Dissert. I. 3, 3.

* Marc. Aur. Med. II. 2 ; XII. 3.

5 Senec. Ep. 65. 24;Epict. Diss. II. 8, 2 ; Marc. Aur. Med. XII. 2.

6 Claud. Mam. De Stratu Arnim. II. 7.

7 In so far as here, too, man is regarded as a microcosm. Ps.-Pythag. iijt

Phot. Cod. 249, p. 440 a.

* Apollonius ol Tyana (neol dvaiQv) in Eus. Prcep. Ev. IV. 13.

Page 251: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 19.] Spirit and Matter : Philo, Plutarch. 231

We meet this same conception ethically, among the Essenes, andtheoretically, everywhere in the teaching of Philo. He, too, dis-

tinguishes between the soul, which as vital force of the bodily

organism has its seat in the blood, and the pneuma, which as ema-

nation of the purely spiritual deity, constitutes the true essential

nature of man. 1 He, too, finds that this latter is imprisoned in the

body, and retarded in its unfolding by the body's sensuous nature

(ai.'o-0£o-is), so that since man's universal sinfulness 2is rooted in this,

salvation from this sinfulness must be sought only in the extirpa-

tion of all sensuous desires ; for him, too, matter is therefore the

corporeal substratum, which has indeed been arranged by the deity

so as to form the purposive, good world, but which, at the same

time, has remained the ground of evil and of imperfection.

2. The Christian Apologists' idea is related to this and yet differ-

ent. With them the Aristotelian conception of God as pure intel-

lect or spirit (vovs i-eAeios) is united with the doctrine that God has

created the world out of shapeless matter : yet here matter is not

regarded immediately as an independent principle, but the ground

of evil is sought rather in the perverted use of freedom on the part

of man and of the demons who seduce him. Here the ethical and

religious character of the dualism of the time appears in its com-

plete purity : matter itself is regarded as something of an indiffer-

ent nature, which becomes good or evil only through its use by

spiritual powers. In the same manner Hellenistic Platonists like

Plutarch, proceeding from the conception of matter as formless Not-

being, sought the principle of evil not in it, but rather in a force or

power, standing in opposition to the good deity,3— a force which,

to a certain degree, contends with the deity about the formation of

matter. Plutarch found this thought in the myths of different

religions, but he might also have referred to a passage where Plato

had spoken of the evil world-soul in opposition to the good. 4

Meanwhile, the tendency to identify the antithesis of good and

evil with that of mind (or spirit) and matter asserts itself here too,

in the fact that the essence of evil is sought again in a propensity

1 In this connection Philo calls -rrvev/xa that which among the Stoics, Aristo-

telians, and Platonists of the time is called vovs ; cf. Zeller V.3 395, 3. Yet there

occur with him again other expressions in which, quite in the Stoic fashion, the

pneuma appears as air, in the sense of a most refined physical reality. Cf. H.Siebeck, Gesch. d. Psych. I. b 302 ff.

8 It is also characteristic that the sinfulness of all men, a doctrine whichis completely at variance with the old Stoic faith in the realisation of the ideal

of the wise man, is generally acknowledged by the Stoics of the time of the

Empire, and regarded as motive for the necessity of supernatural help. Cf.

Seneca, Benef. I. 10 ; VII. 27 ; Epict. Dissert. II. 11, 1.

spiut. De Md. 46 ff.

* Plat. Laws, 896 B.

Page 252: Windelband, History of Philosophy, Vol. 1

232 Hellenistic-Roman Thought : Religious Period. [Part II

toward the sensuous and fleshly, — toward matter ; while the good,

on the contrary, is sought in love to the purely spiritual deity.

This is not only a fundamental feature of the early Christian morals,

but it is found also, in the same form, among the Platonists above

mentioned. For Plutarch, too, liberation from the body is the

necessary preparation for that reception of the working of divine

grace which forms the goal of human life, and when Numenius

carried out his theory further, by teaching that, as in the universe,

so also in man, two souls, one good and one evil, contend with each

other, 1 he yet also seeks the seat of the evil soul in the body and

its desires.

In these doctrines, also, we find everywhere emphasised, not only

the pure spirituality and incorporeality of God, but likewise the

incorporeality of the individual spirit or mind. With Plutarch this

is shown once more in the form that he would separate the vovs, the

rational spirit, from the tf/vxr/, which possesses the sensuous nature

and the passions together with the power to move the body. So, too,

Irenceus 2 distinguishes the psychical breath of life (wor) ^wj/s)

which is of a temporal nature and bound to the body, from the ani-

mating spirit (irvtvfxa £,uioiroiovv) , which is in its nature eternal.

These views of course appear everywhere in connection with the

doctrines of immortality or of the pre-existence and transmigration

of souls, of the Fall through which or as a punishment for which

man has been placed in matter, and of the purification through which

he is to free himself from it again ; and just in this, too, the synthe-

sis in question is completed more and more effectively, inasmuch as

the immutable Eternal which remains ever the same (the Platonic

owcrta) is recognised in spirit; the perishable and changeable in matter.

3. In these connections we find developing gradually a separa-

tion of the two characteristics which had been originally united

in the conception of the soul, — the physiological and the psycholog-

ical, the characteristic of vital force and that of the activity of con-

sciousness. As in the scheme that had already been employed by

Aristotle, so now, side by side with the "soul" which moves the body,

appears the " spirit " as self-subsisting and independent principle,

and in this spirit is found no longer merely a general rational activ-

ity, but the proper essence of the individual (as also of the divine)

personality. The triple division of man into body, soul, and spirit

is introduced m all lines, in the most various modes of expression,3

1 Jambi. m Stob. Eel. I. 894.2 Iren. Adv. Haer. V. 12, 2.3 Of the various terminology (^vxv, anitna, Trvev/xa, spmtus, animus, etc.), in

which these doctrines appear, examples have alreadv been given above, and

Page 253: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 19.] Spirit and Matter: Plotinus. 233

and it is easily understood that in this case, the boundaries, on the

one hand between soul and body, and on the other to a still greater

degree between soul and spirit, were very fluctuating ; for the soul

plays here the part of a mean between the two extremes, matter andspirit.

An immediate consequence of this was that a new and deeper idea

could be gained of the activities of consciousness, which now as

"mental" or "spiritual" were separated from the physiological

functions of the soul. For, when once removed in essence from the

corporeal world, the spirit could not be thought as dependent uponsensuous influences, either in its activity or in the object of its

activity ; and while, in all Greek philosophy, cognition had been

regarded as the perception and taking up of something given, and

the attitude of thought as essentially receptive, now the idea of

mind or spirit as an independent, productive principle forces its

way through.

4. The beginnings for this lie already in the Neo-Pythagorean

doctrine, in so far as in it the spirituality of the immaterial ivorld

was first maintained. The immaterial substances of Platonic meta-

physics, the Ideas, appear no longer as self-subsistent essences, but

as elements constituting the content of intellectual or spiritual activity

;

and while they still remain for human cognition something given

and determining, they become original thoughts of God. 1 Thus the

bodiless archetypes of the world of experience are taken up into

the inward nature of mind ; reason is no longer merely something

which belongs to the ovo-ia or which is only akin to it, it is the

entire ovaia itself ; the immaterial ivorld is recognised as the ivorld of

mind or spirit.2

In correspondence with this, the rational spirit or intellect (vovs)

is defined by Plotinus 3 as the unity which has plurality within

itself, i.e. in metaphysical language, as duality determined by unity

but in itself indeterminate (cf. § 20), and in anthropological lan-

might very easily be multiplied. This doctrine was developed in an especially

interesting way by Origen (De Princ. III. 1-5), where the "soul" is treated

partly as motive power, partly as faculty of ideation and desire, while the spirit,

on the contrary, is presented as the principle of judging, on the one handbetween good and evil, on the other hand between true and false; in this alone,

teaches Origen, consists man's freedom. The like triple division appears then

with Plotinus in connection with his whole metaphysical construction. Enn.II. 9, 2. Cf. § 20.

1 Cf. Nicomachus, Arithm. Intr. I. 6.2 With this change the Platonic doctrine of Ideas passed over to the future,

because Plotinus, and with him all Neo-Platonism, accepted it. Yet. this did not

take place without opposition. Longinus at least protested against it, and Por-

phyry as his disciple wrote a treatise of his own On e£u> rod pol v<pi<TT^K€ t& vot)t&.

Porph. Vit. Plot. 18 ff.

8 Plot. Enn. V. 9, 6 ; 3, 15 ; 4, 2.

Page 254: Windelband, History of Philosophy, Vol. 1

234 Hellenistic-Roman Thought: Religious Period. [Part II.

guage, as the synthetic function which produces plurality out of its

higher unity. From this general point of view the Neo-Platonists

carried out the psychology of cognition under the principle of the

activity of consciousness. For according to this, the higher soul can

no longer be looked upon as passive, but must be regarded as essen-

tially active in all its functions. 1 All its intelligence (crwecris) rests

upon the synthesis (crwöeo-is) of various elements; 2 even where the

cognition refers to what is given by the senses, it is only the body

which is passive, while the soul in becoming conscious (crwauröeo-is

and TrapaKoXovdrjo-is) is active

;

3 and the same is true of the sensuous

feelings and passions. Thus in the field of sensation a distinction

is made between the state of excitation and the conscious perception

of this ; the former is a passive or receptive state of the body (or

also of the lower soul); the latter even already in conscious per-

ception (dvTiATji/'is) is an act of the higher soul, which Plotinus

describes as a kind of bending back of thought— reflection. 4

While consciousness was thus conceived as the active noting of the

mind's own states, functions, and contents,— a theory, which, ac-

cording to Philoponus, was carried out especially by the Neo-Pla-

tonic Plutarch also,— there resulted from this with Plotinus the

conception of self-consciousness (Trapa.Ko\ov9eiv cavrw). 5 His conception

of this was that the intellect, as thought active and in motion

(vo^o-ts), has for its object itself as a resting, objective thought

(votjtov) : intellect as knowledge, and intellect as Being, are in this

case identical.

But the conception of self-consciousness takes on also an ethico-

religious colouring in accordance with the thought of the time. The

o-wecrts is at the same time o-wet'S^o-is— conscience, i.e. man's knowl-

edge, not only of his own states and acts, but also of their ethical

worth, and of the commandment by the fulfilment of which the

estimate of this worth is governed ; and for this reason the doctrine

of self-consciousness is developed in the doctrine of the Church

Fathers, not only as man's knowledge of his sins, but also as repent-

ance (//.eravoia) in actively combating them.

5. The conception of mind or spirit as self-active, creative principle

did not stop with its significance for psychology, ethics, and theory

1 Porph. Sentent. 10, 19 et al.2 Plot. Enn. IV. 3, 26.3 lb. IV. 4, 18 f. The term ffvvala-$T]<ns— whose meaning reminds us besides

of the Koivbv al<ydt]T-f)pwv in Aristotle, and thus ultimately of Plato, Thecet. 184 f.

— is found in similar use already in Alexander Aphrodisias, Qacest. III. 7,

p. 177, and so, too, Galen employs the expression didyvueis to designate thebecoming conscious of the change in the bodily organ as contrasted with that

change itself.

4 Plot. Enn. I. 4, 10. 5 lb. III. 9.

Page 255: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 20.] Grod and the World. 235

of knowledge, but as the ancient world passed out, this conception

rose to be the dominant thought of religious metaphysics. For by

making the attempt to derive matter also from this creative spirit,

this conception offered the possibility of finally overcoming that

dualism which formed the presupposition of the whole movement of

the religious thought of the time.

Hence it became the last and highest problem of ancient philoso-

phy to understand the world as a product of spirit, to comprehend

even the corporeal world with all of its phenomena as essentially

intellectual or spiritual in its origin and content. The spiritualisa-

tioii of the universe is the final result of ancient philosophy.

Christianity and Neo-Platonism, Origen and Plotinus, alike

worked at this problem. The dualism of spirit and matter remains,

indeed, persisting in full force for both so far as they have to do

with the conception of the phenomenal world, and especially when

they treat ethical questions. The sensuous is still regarded as that

which is evil and alien to God, from which the soul must free itself

in order to return to unity with pure spirit. But even this dark

spot is to be illumined from the eternal light, matter is to be recog-

nised as a creation of spirit. The last standpoint of ancient philos-

ophy is thus spiritual monism.

But in the solution of this common problem the philosophy of

Christianity and that of JSTeo-Platonism diverge widely ; for this de-

velopment of the divine spirit into the world of phenomena, even

down to its material forms, must evidently be determined by the

ideas which obtained of the nature of God and of his relation to the

world, and just in this Hellenism found itself working under pre-

suppositions that were completely different from those of the doctrine

of the new religion.

§ 20. God and the World.

The peculiar suspense between metaphysical monism and ethico-

religious dualism, which defines the character of the entire Alex-

andrian philosophy, forces together all the thoughts of the time,

and condenses them into the most difficult of problems, that of the

relation of God and the World.

1. This problem had already been suggested from the purely

theoretical side, by the opposition between the Aristotelian and

the Stoic philosophy. The former maintained the transcendence

of God, i.e. his complete separation from the world, as strongly as

the latter maintained the immanence of God, i.e. the doctrine that

God is completely merged in the world. The problem, and the

fundamental tendency adopted in its solution, may, therefore, be

Page 256: Windelband, History of Philosophy, Vol. 1

236 Hellenistic-Roman Thought : Religious Period. [Fart II

recognised already in the eclectic mingling 1 of Peripatetic and Stoic

cosmology, as type of which the psendo-Aristotelian treatise, Con-

cerning the World is regarded.2 With the Aristotelian doctrine

that the essence of God must be set far above Nature (as the sum-

total of ail particular things which are moved), and especially above

the mutation of earthly existence, is connected here the Stoic en-

deavour to follow the working of the divine power through the entire

universe, even into every detail. While, accordingly, the world was

regarded among the Stoics as God himself, while Aristotle saw in

it a living being, purposefully moved, whose outermost spheres

were set in revolution only by longing for the eternally unmoved,

pure Form, — a revolution communicating itself with ever-lessening

perfection to the lower spheres, — here the macrocosm appears as

the system of individual things existing in relations of mutual

sympathy, in which the power of the supra-mundane God is domi-

nant under the most varied forms as the principle of life. The

mediation between theism and pantheism is gained, partly by the

distinction between the essence and the power of God, partly by

the graded scale of the divine workings, which descends from the

heaven of the fixed stars to the earth. The pneuma doctrine is

united with the Aristotelian conception of God, by conceiving of

the forces of Nature's life as the workings of pure Spirit.3

This turn, however, but increased the difficulty already inherent

in the Aristotelian doctrine of the action of the deity upon the

world. For this action was regarded as consisting in the motion of

matter, and it was hard to reconcile this materialisation of the

divine action with the pure spirituality which was to constitute the

essence of the deity. Even Aristotle had not become clear as to the

relation of the unmoved mover to that which was moved (cf. § 13. ).4

2. The problem became more severe as the religious dualism

became more pronounced, a dualism which, not satisfied with con-

trasting God as spirit with matter, the supersensuous sphere with

the sensuous, rather followed the tendency to raise the divine being

1 Stratonism as a transformation of the Aristotelian doctrine in the direction

of pantheistic immanence, a transformation allied to the doctrine of the Stoa,

has been treated above, § 15, 1.

2 This book (printed among the writings of Aristotle, 391 ff.) may perhapshave arisen in the first century a.d. Apuleius worked it over into Latin.

3 Cf. principally Ch. 6, 397 b 9.4 These difficulties in Aristotle's case became condensed in the concept of the

d<pri. For since the "contact " of the mover with the moved was regarded as the

condition of motion, it was necessary to speak also of a " contact " between Godand the heaven of the fixed stars. This, however, was liable to objection onaccount of the purely spiritual essence of the deity, and the dcprj in this case

received a restricted and intellectually transformed meaning ("immediaterelation "). Cf. Arist. De Gen. et Corr. I. 6, 323 a 20.

Page 257: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 20.] God and the World : Neo-Platonism. 231

above all that can be experienced and above every definite content,

and thus to make the God who is above the world also a God above

mind or spirit. This is found already with the Neo-Pythagoreans,

among whom a wavering between various stadia of dualism lurks

behind their mode of expression in the symbolism of numbers.

When the "One" and the "indefinite duality" are maintained to be

principles, the latter indeed always means matter as the impure, as

the ground of the imperfect and the evil; the One, however, is

treated now as pure Form, as spirit, now also as the "cause of

causes" which lies above all reason, — as the primordial being

which has caused to proceed forth from itself the opposition of the

derivative One and duality, of spirit and matter. In this case the

second One, the first-born One (-n-payTOyovov h) appears as the perfect

image of the highest One. 1

Inasmuch as mind or spirit was thus made a product of the deity,

though the first and most perfect product, this effort led to raising

the conception of the deity even to complete absence of all qualities.

This had been already shown in Philo, who emphasised so sharply

the contrast between God and everything finite that he designated

God expressly as devoid of qualities (a7rotos2) : for since God is

exalted above all, it can be said of him only that he has none of the

finite predicates known to human intelligence ; no name names him.

This type of thought, later called " negative theology," we find also

among those Christian Apologists that were influenced in their con-

ceptions by Philo, especially with Justin,3 and likewise in part

among the Gnostics.

The same meets us also in Neo-Platonism in a still more intensi-

fied form, if possible. As in the Hermetic writings 4 God had been

considered as infinite and incomprehensible, as nameless, exalted

above all Being, as the ground of Being and Reason, neither of

which exists until created by him, so for Plotinus, the deity is the

absolutely transcendent primordial being, exalted as a perfect unity

above mind, which, as the principle that contains plurality already

in its unity (§ 19, 4), must have proceeded forth from God (and

not have been eternal). This One, to ev, precedes all thought and

Being; it is infinite, formless, and "beyond" (eVe'/ceiva) the intel-

lectual as well as the sensuous world, and therefore without con-

sciousness and without activity.5

1 Nicomachus, Theol. Arithm. p. 44.2 Phil. Leg. Aller/. 47 a

;Qu. D. S. Immut. 301 a.

3 Just. Apol. I. Gl ff.4 Poemand. 4 f.

5 It is easy to understand how a state of ecstasy devoid of will and conscious-ness and raised above reason, appeared requisite for man's relation to this supra-rational God-Being, exalted above all actinn, will, and thought. Cf. above, § 18, 0.

Page 258: Windelband, History of Philosophy, Vol. 1

238 Hellenistic-Roman Thought : Religious Period. [Part II.

Finally, while Plotmus still designates this inexpressible First

(to 7rpwTov) as the One, which is the cause of all thought and of all

Being, and as the Good, as the absolute end of all that comes to

pass, even this did not satisfy the later members of the school.

Jamblichus set above the h of Plotinus a still higher, completely

ineffable One (iravrr] app^Tos äpxv *)> and Proclus followed him in this.

3. In opposition to such dialectical subtilisations, the development

of Christian thought in the Church preserved its impressive energy

by holding fast to the conception of God as spiritual personality. It

did this, not as the result of philosophical reflection and reasoning,

but by virtue of its immediate attachment to the living belief of the

Church community, and just in this consisted its psychological

strength, its power in the world's history. This faith is breathed in

the New Testament ; this is defended by all the supporters of

patristic theology, and just by this are the limits of the Christian

doctrine everywhere defined, as against the Hellenistic solutions of

the chief problem in the philosophy of religion.

Hellenism sees in personality, in however purely spiritual a man-

ner it may be conceived, a restriction and a characteristic of the

finite, which it would keep at a distance from the Supreme Being,

and admit only for the particular gods. Christianity, as a living

religion, demands a personal relation of man to the ground of the

world conceived of as supreme personality, and it expresses this

demand in the thought of the divine sonship of man.

If, therefore, the conception of personality as intrinsic spiritual-

ity (geistiger Innerlichkeit) expresses the essentially new result, to

yield which, theoretical and ethical motives intertwined in Greek

and Hellenistic thought, then it was Christianity which entered

upon this inheritance of ancient thought, while Neo-Platonism

turner"1

, back to the old idea that saw in personality only a transi-

tory product of a life which as a whole is impersonal. It is the

essential feature of the Christian conception of the world that it

regards the person and the relations of persons to one another as

the essence of reality.

4. In spite of this important difference, all lines of the Alexan-

drian philosophy were confronted by the same problem, that of plac-

ing the deity, thus taken from the sensible world, in those relations

which religious need demanded. For the more deeply the opposi-

tion between God and the world was felt, the more ardent became

the longing to overcome it — to overcome it by a knowledge that

should understand the world also through God, and by a life that

should return out of the world to God.

1 Damasc. De Princ. 43.

Page 259: Windelband, History of Philosophy, Vol. 1

Chap 2, § 20.] God and the World: Christianity, Gnostics. 239

Hence the dualism of God and the world, as well as that of spirit

and matter, is but the starting-point — taken in the feelings — andthe presupposition of the Alexandrian philosophy: its goal is

everywhere, theoretically as well as practically, to vanquish this

dualism. Just in this consists the peculiarity of this period, that

it is anxious to close, in knowledge and will, the cleft which it finds

n its feelings.

This period, to be sure, produced also theories of the world in

which dualism asserted itself so predominantly as to become fixed

as their immovable basis. Here belong primarily Platonists like

Plutarch, who not only treated matter as an original principle side

by side with the deity, because the deity could in nowise be the

ground of the evil, but also assumed beside God, the " evil world-

soul " as a third principle in the formation of this indifferent matter

into a world. A part of the Gnostic systems present themselves

here, however, for especial consideration.

This first fantastic attempt at a Christian theology was ruled

throughout by the thoughts of sin and redemption, and the funda-

mental character of Gnosticism consists in this, that from the point

of view of these ruling thoughts the conceptions of Greek philos-

ophy were put in relation with the myths of Oriental religions.

Thus with Valentinus, side by side with the deity (nponaTaip) poured

out into the Pleroma or fulness (to 7rXrjpwp.a) of spiritual forms,

appears the Void (to kwu/aol), likewise original and from eternity;

beside Form appears matter, beside the good appears the evil, and

though from the self-unfolding of the deity (cf. 6, below) an entire

spiritual world has been formed in the " fulness " above men-

tioned, the corporeal world is yet regarded as the work of a fallen

Mon (cf. § 21) who builds his inner nature into matter. So, too,

Saturninus set matter, as the domain of Satan, over against God's

realm of light, and regarded the earthly world as a contested bound-

ary province for whose possession the good and evil spirits strive

by their action upon man ; and in a similar manner the mythology

of Bardesanes was arranged, which placed beside the " Father of

Life " a female deity as the receptive power in the formation of the

world.

But dualism reached its culmination in a mixed religion which

arose in the third century under the influence of the Gnostic systems

combined with a return to the old Persian mythology,

Manir

chceism. 1 The two realms of good and evil, of light and darkness,

1 The founder, Mani (probably 240-280 a.d.). regarded his doctrine as th<«

consummation of Christianity and as a revelation of the Paraclete. He fell

ft victim to the persecution of the Persian priests, but his religion soon became

Page 260: Windelband, History of Philosophy, Vol. 1

240 Hellenistic-Roman Thought : Religious Period. [Part II

of peace and strife, stand here opposed as eternally as their princes,

God and Satan. Here, too, the formation of the world is conceived of

as a mixture of good and evil elements,— brought about by a viola-

tion of the boundaries ; in man the conflict of a good soul belonging

to the realm of light, and of an evil soul arising from darkness, is

assumed, and a redemption is expected that shall completely sepa-

rate both realms again.

Thus at the close of the period it is shown in the clearest manner

that the dualism of the time rested essentially upon ethico-religious

motives. By adopting as their point of view for theoretical explana-

tion the judgment of worth, in accordance with which men, things,

and relations are characterised as good or bad, these thinkers came

to trace the origin of the thus divided universe back to two different

causes. In the proper sense of the judgment, only one of these

causes, that of the good, should be regarded as positive and have

the name of deity, but in a theoretical aspect the other also fully

maintains its claim to metaphysical originality and eternity (oiaia).

But even from this relation it may be seen that as soon as the meta-

physical relation was completely adapted to the ethical, this must in

itself lead to a removal of the dualism.

5. In fact, dualism, from motives that were most peculiarly its

own, produced a series of ideas through which it prepared its ownovercoming. For the sharper the antithesis between the spiritual

God and the material world, and the greater the distance between

man and the object of his religious longing, the more the need

asserted itself of bringing about again, by intermediate links, a union

of what was thus separated. The theoretical significance of this

was to render comprehensible and free from objections the action

of the deity upon matter alien to him and unworthy of him;prac-

tically these links had the significance of serving as mediators

between man and God, having the power to lead man out of his sen-

suous vileness to the Supreme Being. Both interests were alike

suggestive of the methods by which the Stoics had known how to

utilise, in their religion of Nature, the popular faith in the lower

deities.

This mediation theory was first attempted on a large and thorough

plan by Philo, who gave it its definite direction by bringing it into

close relations, on the one hand, with the Neo-Pythagorean doctrine

of Ideas, on the other hand with the doctrine of angels in his

greatly extended, and maintained itself in vigour far on into the Middle Ages.We are best instructed with regard to it through Augustine, who was himself

for a time an adherent of it. Cf. F. C. Baur, Das manichäische Religions-

system (Tübingen, 1836); O. Flügel, Mani und seine Lehre (Leins. 1862).

Page 261: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 20.] God and the World : Philo. 241

religion. The mediating powers, in considering which Philo had in

mind more the theoretical significance and the explanation of the

influence of God upon the world, he designates according to the

changing point of view of his investigation, now as Ideas, now as

acting forces, or again as the angels of God; but with this is always

connected the thought that these intermediate members have part

in God as in the world, that they belong to God ami yet are different

from him. So the Ideas are regarded, on the one hand, in Neo-

Pythagorean fashion as thoughts of God and content of his wis-

dom, but again, after the old Platonic thought, as an intelligible

world of archetypes, created by God : and if these archetypes are

held to be at the same time the active forces which shape the unor-

dered matter according to their purposeful meaning, the forces

appear in this case sometimes as powers so independent that by

assigning them the formation and preservation of the world, all

immediate relation between God and the world is avoided, and some-

times again as something attached to the divine essence and repre-

senting it. Finally, as angels they are indeed real mythical forms,

and are designated as the servants, the ambassadors, the messengers,

of God, but on the other hand they represent the different sides and

qualities of the divine essence, which, it is true, is as a whole un-

knowable and inexpressible in its depth, but which reveals itself

just in them. This double nature, conditioned by the fundamental

thought of the system itself, brings with it the consequence that

these ideal forces have the significance of the contents of general

conceptions, and yet are at the same time furnished with all the

marks of personality ; and just this peculiar amalgamation of scien-

tific and mythical modes of thought, this indefinite twilight in which

the entire doctrine remains, is the essential and important therein.

The same is true of the last inference, with which Philo con-

cluded this line of thought. The fulness of Ideas, forces, and

angels was itself in turn an entire world, in which plurality and

motion ruled : between it and the one unmoved, changeless deity

there was need of still a higher intermediate link. As the Idea is

related to the individual phenomena, so the highest of the Ideas

(to yei'iKcuTaTov), the "Idea of the Ideas," must be related to the

Ideas themselves, — as force is related to its activities in the world

of sense, so the rational World-force in general must be related to

the forces : the world of angels must find its unitary conclusion in

an archangel. This sum-total of the divine activity in the world.

Philo designates by the Stoic conception of the Logos. This also

appears with him, on this account, in wavering, changing light.

The Logos is, on the one hand, the divine wisdom, resting within

Page 262: Windelband, History of Philosophy, Vol. 1

242 Hellenistic-Roman Thought : Religious Period. [Part II«

itself (aocfiia.— Adyos evSiaöeros ; cf. p. 200, note 1), and the producing

rational power of the Supreme Being; it is, on the other hand,

Reason as coming forth from the deity (Aoyos 7rpo<^o,otKOs, " uttered

Reason"), the self-subsistent image, the tirst-born son, who is not,

as is God, without origin, nor yet has he arisen, as have we men;

he is the second God. 1 Through him God formed the world, and he

is in turn also the high priest, who, through his intercession, creates

and preserves relations between man and the deity. He is know-

able, while God himself, as exalted above all determination, remains

unknowable : he is God in so far as God forms the life-principle

of the world.

Thus the transcendence and immanence of God divide as separate

potencies, to remain united, nevertheless ; the Logos, as the Godwithin the world, is the " dwelling-place" of the God without the

world. The more difficult the form which this relation assumes

for abstract thought, the richer the imagery in which it is set forth

by Philo. 2

6. With this Logos doctrine the first step was taken toward

filling the cleft between God and the sensible world by a definite

graded succession of forms, descending, with gradual transitions,

from unity to plurality, from unchangeableness to changeableness,

from the immaterial to the material, from the spiritual to the sen-

suous, from the perfect to the imperfect, from the good to the bad;

and when this series, thus arranged by rank, was conceived of at

the same time as a system of causes and effects which again were

themselves causes, there resulted from this a new exposition of the

cosmogonic process, in which the world of sense was derived from

the divine essence by means of all these intermediate members.

At the same time, the other thought was not far distant, that the

stages of this process should be regarded also in their reverse order,

as the stages by which man, ensnared in the world of sense, becomes

reunited with God. And so, both theoretically and practically, the

path is broken on which dualism is to be overcome.

A problem was thus taken up again which Plato in his latest

Pythagoreanising period had had in mind, and the oldest Academi-

cians as well, when they sought, with the aid of the number theory,

1 Philo in Eus. Prcep. Ev. VII. 13, 1. With a somewhat stronger emphasisupon personality, these same conceptions are found in Justin., Apol. I. 32 ; Dial,

c. Tryph. 56 f.

2 Connected with all these doctrines is the fact, that with Philo the spiritual

in the world of experience occupies a doubtful position between the immaterialand the material: the vovs of man, the faculty of thought and will, is a part of

the divine Logos (even the demons are designated after the Stoic analogy as

A670O, and yet it is again characterised as finest pneuma.

Page 263: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 20,] God and the World : Gnostics. 243

to comprehend how Ideas and things proceeded forth from the

divine unity. But it had been shown at that time that this schemeof the development of plurality out of the One, as regards its

relation to the predicates of worth, admitted two opposite interpre-

tations : viz. the Platonic mode of view, defended by Xenocrates,

that the One is the good and the perfect, and that that which is

derived from this is the imperfect and, ultimately, the bad, and the

opposing theory, held by Speusippus, that the good is only the final

product, not the starting-point of the development, and that this

starting-point is to be sought, on the contrary, in the indefinite, the

incomplete. 1 It is customary to distinguish the above-described

doctrines as the system of emanation and the system of evolution.

The former term arises from the fact that in this system, which was

decidedly prevalent in the religious philosophy of Alexandrianism,

the separate formations of the world-producing Logos were often

designated by the Stoic term, as " emanations " (d7roopoiai) of the

divine essence.

Yet the Alexandrian philosophy is not lacking in attempts at

evolutionary systems. In particular, these were especially avail-

able for Gnosticism; for, in consequence of the degree to which it

had strained the dualism of spirit and matter, this system was

necessarily inclined to seek the monistic way of escape rather in an

indifferent, original ground, which divided itself into the opposites.

Hence where the Gnostics sought to transcend dualkm, — and this

was the case with the most important of them,— they projected

not only a cosmogonic but a theogonic process, by which the deity

unfolded himself from the darkness of his primeval essence,

through opposition, to complete revelation. Thus, with Basileides,

the nameless, original ground is called the not (yet) existing God

(6 ovk wv #eds). This being, we hear, produced the world-seed

(7ra>/<nrepiiia), in which the spiritual forces (uiot^tcs) lay unordered

side by side with the material forces (d^opcpta) . The forming and

ordering of this chaos of forces is completed by their longing for the

deity. In connection with this process the various "sonships," the

spiritual world {yirepKoafxta), separate themselves from the material

world (koo>ios), and in the course of the process of generation all the

spheres of the thus developed deity ultimately become separate;

each attains its allotted place, the unrest of striving ceases, and the

peace of glorification rests over the All.

Motives from both systems, that of evolution and that of emana-

tion, appear peculiarly mingled in the doctrine of Valentin us. For

i Cf. Arist. Met. XIV. 4, 1091 b 16; XII. 7. 1072 b 31.

Page 264: Windelband, History of Philosophy, Vol. 1

244 Hellenistic-Roman Thought : Religious Period. [Fart II

here the spiritual world (irXrjpwfjLa) or system of the "JEons," the

eternal essences, is developed first as an unfolding of the dark and

mysterious primitive Depth (ßvöos) to self-revelation, and in the

second place as a descending production of more imperfect forms.

The mythical schema in this is the Oriental pairing of male and

female deities. In the highest pair or " syzygy " there appears side

by side with the original Ground " Silence " (tny^ ) , which is also called

"Thought " (hvoia). From this union of the Original Being with

the capacity of becoming conscious there proceeds as the firstborn the

Spirit (here called voD?) which in the second syzygy has as its object

" Truth," i.e. the intelligible world, the realm of Ideas. Thus, having

itself come to full revelation, the deity in the third syzygy takes the

form of "Reason" (Aöyos) and "Life" (£w?;), and in the fourth

syzygy becomes the principle of external revelation as " Ideal Man "

(av0pw7ros) and "Community" (eK/cA^om, church). While the de-

scending process has thus already begun, it is continued still farther

by the fact that from the third and fourth syzygies still other ^Eons

proceed, which, together with the sacred Eight, form the entire

Pleroma, but which stand farther and farther removed from the

original Ground. It is the last of these ^Eons, " Wisdom "(oro<£uz),

that, by sinful longing after the original Ground, gives occasion for

the separation of this Longing and of its being cast into the mate-

rial Void, the kwu/m, there to lead to the formation of the earthly

world.

If we look at the philosophical thoughts which lie back of these

highly ambiguous myth-constructions, it is easy to understand that

the school of the Valentinians diverged into various theories. For

in no other system of that time are dualistic and monistic motives

of both kinds, from the system of evolution as well a.° from that of

emanation, so intricately mingled.

7. Clarified conceptionally, and freed from mythical apparatus,

the like motives appear in the doctrine of Plotinns, yet in such a

manner that in the system as completed the principle of emanation

almost entirely crowds out the other two.

The synthesis of transcendence and immanence is sought byPlotinus also in the direction of preserving the essence of God as

the absolutely one and unchangeable, while plurality and changea-

bility belong only to his workings. 1 Of the " First," which is ex-

alted above all finite determinations and oppositions, nothing what-

ever can be predicated in the strict sense (cf. above, 2). It is

1 In so far we find here, coined into theological form, the problem of theEleatics and Heraclitus, with which Greek metaphysics began, — a problemwhich also determined the nature of Platonism.

Page 265: Windelband, History of Philosophy, Vol. 1

Cha>\ 2, § 20.] God and the World : Plotinus. 246

only in an improper sense, in its relation to the world, that it can

be designated as the infinite One, as the Good, and as the highest

Power or Force (-n-pwrrj 8vva/xi<;), and the workings of this Power

which constitute the universe are to be regarded, not as ramifica-

tions and parts into which the substance of the First divides, and so

not as " emanations " in the proper sense, but rather as overflowing

by-products which in nowise change the substance itself, even

though they proceed from the necessity of its essence.

To exprass this relation in figurative form Plotinus employs the

analogy of light, — an analogy which, in turn, has also an influence

in determining his conception. Light, without suffering at all in its

own essence or itself entering into motion, shines into the darkness

and produces about itself an atmosphere of brightness that decreases

in intensity more and more from the point which is its source, and

finally of itself loses itself in darkness. So likewise the workings

of the One and Good, as they become more and more separate from

their source, proceeding through the individual spheres, become

more and more imperfect and at last change suddenly into the dark,

evil opposite — matter.

The first sphere of this divine activity is, according to Plotinus,

mind or rational spirit (vovs), in which the sublime unity differen-

tiates itself into the duality of thought and Being, i.e. into that of

consciousness and its objects. In mind the essence of the deity is

preserved as the unity of the thought-function (väqvii) ; for this

thought which is identical with Being is not regarded as an activity

that begins or ceases, changing as it were with its objects, but as the

eternal, pure perception, ever the same, of its own content, which is

of like essence with itself. But this content, the world of Ideas,

the eternal Being {ovcria in the Platonic sense) as contrasted with

phenomena, is, as intelligible world (kÖo><.os vo^tös), at the same time

the principle of plurality. For the Ideas are not merely thoughts

and archetypes, but are at the same time the moving forces (vol

Sum/uis) of lower reality. Because, therefore, unity and variety

are united in this intelligible world as the principles of persistence

and of occurrence and change, and are yet again separated, the fun-

damental conceptions (categories) of this world are these five,1

viz.

Being or Existing (to ov), Rest (crra<ns), Motion or Change (kiVt/o-is),

Identity (tout-ot^s), and Difference (erepÖTr^). Mind, then, as a

function which has determinate contents, and carries plurality

within itself, is the form through which the deity causes all empiri

1 Well known from the dialogue, the Sophist, of the Corpus Platoniatm. CL254 B. ff.

Page 266: Windelband, History of Philosophy, Vol. 1

246 Hellenistic-Roman Thought : Religious Period. [Pakt II

cal reality to proceed forth from itself : God as productive principle,

as ground of the world, is mind or rational spirit.

But spirit needs to shine out in a similar manner in order to pro-

duce the world from itself ; its most immediate product is the soul,

and this in turn evinces its activity by shaping matter into cor-

poreality. The peculiar position of the " soul " therefore consists

in this, that it, perceiving or beholding, receives the content of

spirit, the world of Ideas, and after this archetype (eUw) forms

the world of sense. Contrasted with the creative spirit, it is the

receptive, contrasted with matter, the active principle. And this

duality of the relations toward the higher and the lower is here so

strongly emphasised that just as " spirit " divided into thought and

Being, so the soul, for Plotinus, is out and out doubled : as sunk

into the blissful contemplation of the Ideas it is the higher soul,

the soul proper, the $vxq in the narrower sense of the word; as

formative power, it is the lower soul, the </>wns (equivalent to the

\oyos cnrepfAaTiKos of the Stoics).

All these determinations apply on the one hand to the universal

soul (world-soul — Plato), and on the other to the individual souls

which have proceeded from it as the particular forms which it has

taken on, especially therefore to human souls. The <f>vcn<;, the for-

mative power of Nature, is distinguished from the pure, ideal world-

soul : from the latter emanate the gods, from the former the demons.

Beneath man's knowing soul, which turns back to the spirit, its

home, stands the vital force which forms the body. Thus the sepa-

ration in the characteristics of the concept of the soul— a separation

which developed materially from dualism (cf. § 19, 3) — is here de-

manded formally by the connected whole of the metaphysical system.

In this connection, this working of the soul upon matter is of

course conceived of as purposive, that is, as appropriate or adapted

for ends, because it ultimately goes back to spirit and reason

(Aoyos) ; but since it is a work of the lower soul, it is regarded as

undesigned, unconscious direction, which proceeds according to

natural necessity. As the outer portions of the rays of light pene-

trate into the darkness, so it belongs to the nature of the soul to

illumine matter with its glory which arises from spirit and from

the One.

This matter, however, — and this is one of the most essential

points in the metaphysics of Plotinus, — must not be looked upon

as a corporeal mass subsisting in itself beside the One ; it is, rather,

itself without body, immaterial. 1 Bodies are indeed formed out of

1 do-w/xaros : Ennead. III. 6, 7.

Page 267: Windelband, History of Philosophy, Vol. 1

Chap. "J, § 2U.] God and the World : l*lotinus. 24?

it, but it is itself no body ; and. since it is thus neither spiritual nor

corporeal in its nature, it cannot be determined by any qualities

(d-7rotos). But for Plotinus, this epistemological indeterniinateness

has, at the same time, the force of metaphysical indeterniinateness.

Matter is for him absolute negativity, pure privation {o-Teprjaa),

complete absence of Being, absolute Non-being : it is related to the

One as darkness to light, as the empty to the full. This v\rj of

the Neo-Platonists is not the Aristotelian or the Stoic, but is once

more the Platonic ; it is empty, dark space. 1 So far in ancient

thought does the working of the Eleatic identification of emptyspace with Non-being, and of the farther extension of this doctrine

by Democritus and Plato, extend: in Neo-Platonism, also, space

serves as the presupposition for the multiplication which the Ideas

find in the phenomenal world of sense. For this reason, with

Plotinus, also, the lower soul, or <£wns, whose office it is to shine

out upon matter, is the principle of divisibility, 2 while the higher

soul possesses the indivisibility which is akin to the rational spirit.

In this pure negativity lies a ground for the possibility of deter-

mining by a predicate of worth this matter thus devoid of quali-

ties ; it is the evil. As absolute want (wevta 7ravTeX^s), as the

negation of the One and of Being, it is also the negation of

the Good, cUi-oro-ia ayadov. But by introducing the conception

of evil in this manner, it receives a special form : evil is not itself

something positively existent ; it is want, or deficiency ; it is lack

of the Good, Non-being. This conception thus formed gave Plotinus

a welcome argument for theodicy ; if the evil is not, it need not be

justified, and so it follows from the sheer conceptions as so deter-

mined that all that is, is good.

For Plotinus, therefore, the world of the senses is not in itself

evil any more than it is in itself good; but because in it light

passes over into darkness, because it thus presents a mixture of

Being and Non-being (the Platonic conception of yeVeo-is here comes

into force anew), it is good so far as it has part in God or the

Good; i.e. so far as it is; and on the other hand, it is evil in so far

as it has part in matter or the Evil ; i.e. in so far as it is not [has

no real, positive existence]. Evil proper, the true evil (-n-pdrov

kclkov), is matter, negation; the corporeal world can be called evil

only because it is formed out of matter : it is secondary evil (Seirrepov

«aKov); and the predicate " evil " belongs to souls only if they give

1 Ennead. III. 6, 18. Universal empty space forms the possibility (vwoKelpevov)

for the existence of bodies, while, on the other hand, the particular spatial deter-

minate ness is conditioned by the nature of the bodies, II. 4, 12.2 lb. III. 9, 1.

Page 268: Windelband, History of Philosophy, Vol. 1

248 Hellenistic-Roman Thought : Religious Period. [Part II.

themselves over to matter. To be sure, this entrance into matter

belongs to the essential characteristics of the soul itself ; the soul

forms just that sphere in which the shining forth of the deity

passes over into matter, and this participation in evil is, therefore,

for the soul, a natural necessity which is to be conceived of as

a continuation of its own proceeding forth from the rational

spirit.1

By this distinction of the world of sense from matter, Plotinus

was able to do justice, also, to the positive element in phenomena. 2

For since the original power works through spirit and soul upon

matter, all that in the world of sense really exists or is, is evidently

itself soul and spirit. In this is rooted the spiritualisation of the

corporeal world, the idealising of the universe, which forms the

characteristic element in the conception of Xature held by Plotinus.

The material is but the outer husk, behind which, as the truly

active reality, are souls and spirits. A body or corporeal substance

is the copy or shadow of the Idea which in it has shaped itself to

matter ; its true essence is this spiritual or intellectual element

which appears as a phenomenon in the image seen by sense.

It is in such shining of the ideal essence through its sensuous

phenomenon that beauty consists. By virtue of this streaming of

the spiritual light into matter the entire world of the senses is

beautiful, and likewise the individual thing, formed after its arche-

type. Here in the treatise of Plotinus on beauty {Ennead. I. 6)

this conception meets us for the first time among the fundamental

conceptions of a theory of the world; it is the first attempt at

a metaphysical aesthetics. Hitherto the beautiful had always

appeared only in homonomy with the good and the perfect, and the

mild attempts to separate the conception and make it independent,

which were contained in Plato's Symposium, were now taken up again

for the first time by Plotinus ; for even the theory of art, to which

aesthetic science had restricted itself as it appeared most clearly in

the fragment of the Aristotelian Poetic, considered the beautiful

essentially according to its ethical effects (cf. § 13, 14). Ancient

life must run its entire course, and that turning toward the inner

life, that internalising, as it were, which this life experienced in the

religious period, must be completed, to bring about the scientific

1 Therefore, though Plotinus in his ethics emphasised strongly freedom in

the sense of responsibility, the great tendency of his metaphysical thought is

shown just in this, that he did not make this freedom of "power to the con-trary " his explaining principle, hut sought to understand the transition of theworld into evil as a metaphysical necessity.

2 Very characteristic in this respect is the treatise (Ennead. II. 9) which hewrote against the barbarian contempt of Nature shown by the Gnostics.

Page 269: Windelband, History of Philosophy, Vol. 1

CiiAr. 2, § 20.] God and the World : Plotinus. 249

consciousness of this finest and highest content of the Grecian

world ; and the conception in which this takes place is on this

account characteristic for the development from which it comes

forth; the beauty which the Greeks had created and enjoyed is

now recognised as the victorious power of spirit in externalising its

sensuous phenomena. This conception also is a triumph of the

spirit, which in unfolding its activities has at last apprehended its

own essential nature, and has conceived it as a world-principle.

As regards the phenomenal world, Plotinus takes a point of view

which must be designated as the interpretation of Nature in terms of

psychical life, and so it turns out that with reference to this antithe-

sis ancient thought described its course from one extreme to the

other. The oldest science knew the soul only as one of Nature's

products side by side with many others,— for Neo-Platonism the

whole of Nature is regarded as real only in so far as it is soul.

But by employing this idealistic principle for explaining individ-

ual things and processes in the world of sense, all sobriety andclearness in natural research is at an end. In place of regular,

causal connections appears the mysterious, dreamily unconscious

weaving of the world-soul, the rule of gods and demons, the spirit-

ual sympathy of all things expressing itself in strange relations

among them. All forms of divination, astrology, faith in miracles,

naturally stream into this mode of regarding Nature, and manseems to be surrounded by nothing but higher and mysterious

forces : this world created by spirit, full of souls, embraces him like

a magic circle.

The whole process in which the world proceeds forth from the

deity appears, accordingly, as a timeless, eternal necessity, andchough Plotinus speaks also of a periodical return of the sameparticular formations, the world-process itself is yet for him without

beginning or end. As it belongs to the nature of light to shine

forever into the darkness, so God does not exist without the stream-

ing forth with which he creates the world out of matter.

In this wiiversal life of spirit the individual personality vanishes,

as a subordinate, particular phenomenon. Released from the all-

soul as one of countless forms in which that unfolds, it is cast into

the sensuous body out of the purer pre-existent state, on account

of its guilty inclination toward what is void and vain, and it is

its task to estrange itself from the body and from material essence

in general, and to "purify" itself again from the body. Only whenit has succeeded in this can it hope to traverse backward the stages

by which it has proceeded forth from the deity, and so to return to

the deity. The first positive step to this exaltation is civic and

Page 270: Windelband, History of Philosophy, Vol. 1

250 Hellenistic-Roman Thought : Religious .Period. [Part II

political virtue, by which man asserts himself as a rationally forma*

tive force in the phenomenal world ; but since this virtue evinces

itself only in reference to objects of the senses, the dianoetic virtue

of knowledge stands far above it (cf. Aristotle),— the virtue by which

the soul sinks into its own spiritual intrinsic life. As a help

stimulating to this virtue, Plotinus praises the contemplation of che

beautiful, which finds a presentiment of the Idea in the thing of

sense, and, in overcoming the inclination toward matter, rises from

the sensuously beautiful to the spiritually beautiful. And even

this dianoetic virtue, this aesthetic Oewpla and self-beholding of the

spirit, is only the preliminary stage for that ecstatic rapture with

which the individual, losing all consciousness, enters into unity with

the ground of the world (§ 18, 6). The salvation and the blessed-

ness of the individual is his sinking into the All-One.

The later Neo-Platonists, — Porphyry first, and, still more, Jamblichus andProclus, — in the case of this exaltation emphasise, far more than Plotinus, the

help which the individual finds for it in positive religion and its acts of worship.

For these men largely increased the number of different stages through whichthe world proceeds forth from the " One," and identified them with the formsof the deities in the different ethnic religions by all kinds of more or less arbi-

trary allegories. It was therefore natural, in connection with the return of the

soul to God, since it must traverse the same stages up to the state of ecstatic

deification, to claim the support of these lower gods : and thus as the metaphys-ics of the Neo-Platonists degenerated into mythology, their ethics degeneratedinto theurgic arts.

8. On the whole, therefore, the derivation of the world from God as

set forth by Plotinus, in spite of all its idealising and spiritualising

of Nature, follows the physical schema of natural processes. This

streaming forth of things from the original Power is an eternal

necessity, founded in the essence of this Power ; creation is a pur-

posive working, but unconscious and without design.

But at the same time, a logical motive comes into play here, which

has its origin in the old Platonic character of Ideas as class-concepts.

For just as the Idea is related to individual things of sense, so in

turn the deity is related to Ideas, as the universal to the particular.

God is the absolute universal, and according to a law of formal

logic, in accordance with which concepts become poorer in contents

or intension in proportion as their extension increases so that the

content must correspond to the extension oo, the absolutely uni-

versal is also the concept of the " First," void of all content. BuJ;

if from this First proceed first the intelligible, then the psychical,

and finally the sensuous world, this metaphysical relation corre-

sponds to the logical process of determination or partition. This

point of view, according to which the more general is throughout

regarded as the higher, metaphysically more primitive reality, while

Page 271: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 20.] God and the World : Proclus, Christianity. 251

the particular is held to be, in its metaphysical reality also, a deriv*

ative product from the more general, — a view which resulted from

hypostatising the syllogistic methods of Aristotle (cf. § 12, 3), —was expressed among the older Neo-Platonists principally by

Porphyry, in his exegesis of Aristotle's categories.

Meanwhile Proclus undertook to carry out methodically this

logical schema of emanation, and out of regard for this principle

subordinated a number of simple and likewise unknowable " henads"

beneath the highest, completely characterless h. In so doing he

found himself under the necessity of demanding a proper dialectical

principle for this logical procession of the particular from the uni-

versal. Such a schematism the systematiser of Hellenism found in

the logico-metaphysical relation which Plotinus had laid at the basis

of the development of the world from the deity. The procession of

ihe Many forth from the One involves, in the first place, that the

particular remains like the universal, and thus that the effect abides

or persists within the cause ; in the second place, that this product

is a new self-subsisting entity in contrast with that which has pro-

duced it, and that it proceeds forth from the same ; and finally, that

by virtue of just this antithetic relation the individual strives to

return again to its ground. Persistence, procession, and return (/xovt/,

7rpooSos, eVicn-poc^ ) , or identity, difference and union of that which

has been distinguished, are accordingly the three momenta of the

dialectical process ; and into this formula of emanistic development,

by virtue of which every concept should be thought of as in itself

— out of itself— returning into itself, Proclus pressed his entire

combined metaphysical and mythological construction, — a construc-

tion in which he assigned to the systems of deities of the different

religions their place in the mystical and magical universe, arranging

them in the series divided again and again by threes, according to

his law of the determination of concepts. 1

9. In contrast with this, the peculiarity of Christian philosophy

consists essentially in this, that in its apprehension of the relation

of God to the world, it sought to employ throughout the ethical

point of view of free, creative action. Since from the standpoint of

its religious conviction it held fast to the conception of the person-

ality of the Original Being, it conceived of the procedure of the

world forth from God, not as a physical or logical necessity of the

1 Personally, Proclus is characterised by the mingling of a superabundantcredulous piety with a logical formalism carried even to pedantry, a combina-tion which is highly interesting psychologically. Just for this reason he is,

perhaps, the most pronounced type of this period which is concerned in putting

its ardent religiosity into a scientific system.

Page 272: Windelband, History of Philosophy, Vol. 1

252 Hellenistic-Roman Thought : Religious Period. [Part II..

unfolding of his essence, but as an act of will, and in consequence of

this the creation of the world was regarded not as an eternal process,

but as a fact in time that had occurred once for all. The conception,

however, in which these motives of thought became concentrated,

was that of the freedom of the will.

This conception had had at first the meaning (with Aristotle)

of conceding to the finite personality acting ethically the capacity of

a decision between different given possibilities, independently of

external influence and compulsion. The conception had then taken

on, with Epicurus, the metaphysical meaning of a causeless activity

of individual beings. Applied to the absolute, and regarded as a

quality of God, it is developed in the Christian philosophy into the

thought of " creation out of nothing," into the doctrine of an un-

caused production of the world from the will of God. Every attempt

at an explanation of the world is thereby put aside ; the world is

because God has willed it, and it is such as it is because God has

willed it so to be. At no point is the contrast between JSTeo-Pla-

tonism and orthodox Christianity sharper than at this.

Meanwhile, this same principle of the freedom of the will is

employed to overcome the very difficulties which resulted from it.

For the unlimited creative activity of the omnipotent God forces

the problem of " theodicy " forward still more urgently than in the

other theories of the universe, — the problem how the reality of

evil in the world can be united with God's perfect goodness. Theoptimism involved in the doctrine of creation, and the pessimism in-

volved in the felt need of redemption, the theoretical and the practical,

the metaphysical and the ethical momenta of religious faith strike

hard against each other. But faith, supported by the feeling of

responsibility, finds its way of escape out of these difficulties in the

assumption that God provided the spirits and human souls which

he created, with a freedom analogous to his own, and that through

their guilt evil came into the good world. 1

This guilt, the thinkers of the Church find not to consist properly

in the inclination toward matter or the sensuous; for matter as

created by God cannot in itself be evil. 2 The sin of free spirits

consists rather in their rebellion against the will of God, in their

1 This is expressed abstractly by Clement of Alexandria (Strom. IV. 13, 605)in the form, that evil is only an action, not a substance (ovaia), and that it there-

fore cannot be regarded as the work of God.2 Just for this reason the metaphysical dualism of the Gnostics must be in its

principle heterodox, and that, too, no matter whether it bore the stamp rather

of Oriental mythology or of Hellenistic abstract thought — even though in the

ethical consequences which it drew it coincided in great part with the doctrine

)f the Church.

Page 273: Windelband, History of Philosophy, Vol. 1

Uhap. 2, § 20.] God and the World: Origen. 253

longing after an unlimited power of self-determination, and only

secondarily in the fact that they have turned their love toward God's

creations, toward the world instead of toward God himself. Here

too, therefore, there prevails in the content of the conception of

evil the negative element of departure and falling away from God; '

but the whole earnestness of the religious consciousness asserts

itself in this, that this falling away is conceived of not merely as

absence of the good, but as a positive, perverted act of will.

In accordance with this the dualism of God and the world, and

that of spirit and matter, become indeed deeply involved in the

Christian theory of the world. God and the eternal life of the

spirit, the world and the transitory life of the flesh, — these are

here, too, sharply enough contrasted. In contradiction with the

divine pneuma the world of sense is filled with "hylic" spirits,2

evil demons, who ensnare man in their pursuits which are animated

by hostility to God, stifle in him the voice of universal natural reve-

lation, and thereby make special revelation necessary ; and without

departure from them and from the sensuous nature there is for the

early Christian ethics, also, no rescue of the soul possible.

But still this dualism is not regarded as being in its intrinsic

nature either necessary or original. It is not the opposition be-

tween God and matter, but that between God and fallen spirits ; it

is the purely inner antagonism of the infinite and the finite iv ill. In

this direction Christian philosophy completed through Origen the

metaphysical spiritualising and internalising or idealising of the

world of the senses. In it the corporeal world appears as completely

permeated and maintained by spiritual functions,— yes, even as muchreduced to spiritual functions, as is the case with Plotinus ; but

here the essential element in these functions is relations of ivill.

As the passing over of God into the world is not physical necessity,

but ethical freedom, so the material world is not a last streaming

forth of spirit and soul, but a creation of God for the punishment

and for the overcoming of sin.

To be sure, Origen, in developing these thoughts, took up a motive

which was allied to Neo-Platonism, a motive which brought him

into conflict with the current mode of thought in the Church. For

strongly as he held fast to the conception of the divine personality

and to that of creation as a free act of divine goodness, the scientific

thought which desires to see action grounded in essence was yet

too strong in him to allow him to regard this creation as a causeless

1 In this sense, even Origen could call the evil t6 o'jk 6v (in Joh. II. 7, (35)2 Tatian. Orat. ad Graec. 4.

Page 274: Windelband, History of Philosophy, Vol. 1

254 Hellenistic-Roman Thought : Religious Period. [Part IL

act taking place once for all in time. The eternal, unchangeable

essence of God demands rather the thought that he is creator from

eternity even to all eternity, that he never can be without creating,

that he creates tunelessly. 1

But this creation of the eternal will is, therefore, only one that

relates to eternal Being, to the spiritual world (ovata). In this

eternal manner, so Origen teaches, God begets the eternal Son, the

Logos, as the sum-total of his world-thoughts (i8e'a i8ew), and

through him the realm of free spirits, which,, limited within itself,

surrounds the deity as an ever-living garment. Those of the spirits

that continue in the knowledge and love of the Creator remain in

unchanged blessedness with him ; but those that become weary and

negligent, and turn from him in pride and vainglory, are, for pun-

ishment, cast into matter created for this purpose. So arises the

world of sense, which is, therefore, nothing self-subsistent, but

a symbolic eternalisation of spiritual functions. For what may be

regarded as Real in it is not the individual bodies, but rather the

spiritual Ideas which are present, connected and changing within

them. 2

So, with Origen, Platonism becomes united with the theory of

the creative will. The eternal world of spirits is the eternal prod-

uct of the changeless divine will. The principle of the temporal

and the sensuous (-yeveo-is) is the changing will of the spirits.

Corporeality arises on account of their sin, and will vanish again

with their improvement and purification. Thus will, and the rela-

1 Orig. De Princ. I. 2, 10 ; III. 4, 3.2 This idealising of the world of sense was treated in great detail, quite ac-

cording to the Platonic model, by the most important of the Oriental Churchfathers, Gregory of Nyssa (331-394). His main treatise is the \6yos KaTvxv-tlk6s. Edition of his works by Morellus (Paris, 1675) [Eng. tr. in Vol. V., 2dseries, Lib. Nicene and Post-Nicene Fathers, ed. Schaff and Wace, Oxford,Lond., and N.Y. 1890]. Cf. J. Rupp, G. des Bischofs von N. Leben undMeinungen, Leips. 1834. — This transformation of Nature into psychical termsfound an extremely poetic exposition among the Gnostics, particularly with the

most ingenious among them, Valentinus. The origin of the world of senseis portrayed as follows in his theogonic-cosmogonic poetic invention : Whenthe lowest of the iEons, Wisdom (aocpia), in over-hasty longing, would fain

have plunged into the original Ground and had been brought back again to herplace by the Spirit of Measure (Spos), the Supreme God separated from her herpassionate longing (rrddos) as a lower Wisdom {kAtw aocpia), called Achamoth,and banished it into the "void" (cf. § 20, 4). This lower aocpia, nevertheless,

impregnated by Spos for her redemption, bore the Demiurge and the world of

sense. On this account that ardent longing of aocpia expresses itself in all

forms and shapes of this world ; it is her feelings that constitute the essence of

phenomena ; her pressure and complaint thrills through all the life of Nature.

From her tears have come fountains, streams, and seas ; from her benumbingbefore the divine word, the rocks and mountains ; from her hope of redemption,light and ether, which in reconciliation stretch above the earth. This poetic

invention is farther carried out with the lamentations and penitential songs of

aocpia in the Gnostic treatise, Titans aocpia.

Page 275: Windelband, History of Philosophy, Vol. 1

,*hai\ 2, § 21.] Philosophy of History. 255

Hon of personalities to one another, in particular that of the finite to

the infinite personality, are recognised as the ultimate and deepest

meaning of all reality.

§ 21. The Problem of Universal History.

\Vith this triumph of religious ethics over cosmological meta-

ph}-sics, thus sealed by Christianity, is connected the emergence of

a farther problem, to solve which a number of important attempts

were made — the problem of the philosophy of history.

1. Here something which is in its principle new comes forward,

as over against the Greek view of the world. For Greek science

had from the beginning directed its questions with reference to the

<£u<ris, the abiding essence (cf. p. 73), and this mode of stating

the question, which proceeded from the need of apprehending

Nature, had influenced the progress of forming conceptions so

strongly that the chronological course of events had always been

treated as something of secondary importance, having no meta-

physical interest of its own. In this connection Greek science

regarded not only the individual man, but also the whole humanrace, with all its fortunes, deeds, and experiences, as ultimately but

an episode, a special formation of the world-process which repeats

itself forever according to like laws.

This is expressed with plain grandeur in the cosmological begin-

nings of Greek thought; and even after the anthropological tendency

had obtained the mastery in philosophy the thought remained in

force as theoretical background for every projected plan of the art

of living, that human life, as it has sprung forth from the unchang-

ing process of Nature, must flow again into the same (Stoa). Plato

had indeed asked for an ultimate end of earthly life, and Aristotle

had investigated the regular succession of the forms assumed by

political life ; but the inquiry for a meaning in human history taken

as a ivhole, for a connected plan of historical development, had

never once been put forward, and still less had it occurred to any

of the old thinkers to see in this the intrinsic, essential nature of

the world.

The most characteristic procedure in just this respect is that

of Neo-Platonism. Its metaphysics, also, follows the religious

motive as its guide ; but it gives this motive a genuine Hellenic

turn when it regards the procession of the imperfect forth from

the perfect as an eternal process of a necessary nature, in which

the human individual also finds his place and sees it as his destiny

to seek salvation alone by himself by return to the infinite.

Page 276: Windelband, History of Philosophy, Vol. 1

256 Hellenistic-Roman Thought : Religious Period. [Part II.

2. Christianity, however, found from the beginning the essence

of the whole world-movement in the experiences of jyersonalities

:

for it external nature was but a theatre for the development of

the relation of person to person, and especially of the relation of the

finite spirit to the deity. And to this were added, as a further

determining power, the principle of love, the consciousness of the

solidarity of the human race, the deep conviction of the universal

sinfulness, and the faith in a common redemption. All this led to

regarding the history of the fall and of redemption as the true

metaphysical import of the world's reality, and so instead of an

eternal process of Nature, the drama of universal history as an on-

ward flow of events that were activities of free will, became the con-

tent of Christian metaphysics.

There is perhaps no better proof of the power of the impression

which the personality of Jesus of Nazareth had left, than the fact

that all doctrines of Christianity, however widely they may other-

wise diverge philosophically or mythically, are yet at one in seeking

in him and his appearance the centre of the worloVs history. By him

the conflict between good and evil, between light and darkness, is

decided.

But this consciousness of victory with which Christianity believed

in its Saviour had still another side : to the evil which had been

overcome by him belonged also the other religions, as by no means

its least important element. For the Christian mode of thought of

those days was far from denying the reality of the heathen gods ; it

regarded them rather as evil demons, fallen spirits who had seduced

man and persuaded him to worship them, in order to prevent his

returning to the true God. 1

By this thought the conflict of religions, which took place in the

Alexandi'ian period, acquires in the eyes of Christian thinkers a

metaphysical significance : the powers whose struggling forms the

world's history are the gods of the various religions, and the history

of this conflict is the inner significance of all reality. And since

every individual man with his ethical life-work is implicated in

this great complex process, the importance of individuality becomes

raised far above the life of sense, into the sphere of metaphysical

reality.

3. With almost all Christian thinkers, accordingly, the world's

history appears as a course of inner events which draw after themthe origin and fortunes of the world of sense,— a course which

takes place once for all. It is essential])' only Origen who holds fast

1 So even Origen ; cf. Cont. Cels. III. 28.

Page 277: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 21.] Philosophy of History : Christianity. 257

to the fundamental character of Greek science (cf. p. 27, ch. 1).

so far as to teach the eternity of the world-process. Between the

two motives, the Christian and the Greek, he found a way of escape

by making a succession of temporal worlds proceed forth from the

eternal spiritual world, which he regarded as the immediate creation

of God, and by holding that these temporal worlds take their origin

with the declension and fall of a number of free spirits, and are to

find their end with the redemption and restitution of the same

( u7roKaraCTTacrts ) •

The fundamental tendency of Christian thought, on the contrary,

was to portray the historical drama of fall and redemption as a

connected series of events taking place once for all, which begins

with a free decision of lower spirits to sin, and has its turning-

point in the redemptive revelation, the resolve of divine freedom.

In contrast with the naturalistic conceptions of Greek thought,

history is conceived of as the realm offree acts of personalities, taking

place but once, and the character of these acts, agreeably to the entire

consciousness of the time, is of essentially religious significance.

4. It is highly interesting now to see how in the mythico-

metaphysical inventions of the Gnostics, the peculiar relation of

Christianity to Judaism is brought to expression in cosmogonic

garb. In the Gnostic circles the so-called Gentile Christian ten-

dency is predominant, the tendency which desires to define the newreligion as sharply as possible, as over against Judaism, and this

tendency just through the Hellenistic philosophy grows to the most

open hostility against Judaism.

The mythological form for this is, that the God of the Old Testa-

ment, who gave the Mosaic law, is regarded as the fashioner of the

world of sense,— for the most part under the Platonic name of the

Demiurge,— and is assigned that place in the hierarchy of cosmic

forms or ^Eons, as well as in the history of the universe, whichbelongs to him in accordance with this function.

At the beginning this relation is not yet that of pronounced oppo-

sition. A certain Cerinthus (about 115 a.d.) had already distin-

guished the God of the Jews as Demiurge, from the Supreme Godwho was not defiled by any contact with matter, and had taught

that in contrast with the "law" given by the God of the Jews,

Jesus had brought the revelation of the Supreme God. 2 So, too,

1 Orig. De Princ. III. 1, 3. These worlds, on account of the freedom fromwhich they proceed, are not at all like one another, but are of the most mani-fold variety; lb. II. 3, 3 f.

2 A distinction which Numenius also adopted, evidently under Gnostic influ-

ences. Cf. Euseb. Prcvp. Ev. XI. 18.

Page 278: Windelband, History of Philosophy, Vol. 1

258 Hellenistic-Roman Thouglit : Religious Period. [Part II.

with Saturninus, the God of the Jews appears as the head of the

seven planetary spirits, who, as lowest emanation of the spiritual

realm, in their desire to rule tore away a portion of matter to form

from it the world of sense, and set man as guardian over it. But a

conflict arises, since Satan, to conquer back this part of his kingdom,

sends against man his demons and the lower "hylic" race of men.

In this conflict the prophets of the Demiurge prove powerless until

the Supreme God sends the iEon vovs as Saviour, in order that he

may free pneumatic men and likewise the Demiurge and his spirits

from the power of Satan. This same redemption of the Jewish Godalso is taught by Basilides, who introduces him under the name of

the " great Archon "as an efflux of the divine world-seed, as head

of the world of sense, and represents him as made to tremble by the

Supreme God's message of salvation in Jesus, and as brought to

repentance for his undue exaltation.

In a similar manner, the God of the Old Testament, with Carpo-

crates, belongs to the fallen angels, who, commissioned to form the

world, completed it according to their own caprice, and founded sep-

arate realms in which the}' got themselves reverenced by subordinate

spirits and by men. But while these particular religions are, like

their Gods, in a state of mutual conflict, the Supreme Deity reveals

in Jesus the one true universal religion which has Jesus as its

object, even as he had already before made revelation in the great

educators of humanity, a Pythagoras and a Plato.

In more decided polemic against Judaism Cerdo the Syrian

further distinguished the God of the Old Testament from that of

the New. The God announced by Moses and the prophets, as the

purposeful World-fashioner and as the God of justice is accessible

even to natural knowledge— the Stoic conception ; the God re-

vealed through Jesus is the unknowable, the good God — the

Philonic conception. The same determinations more sharply defined

are employed by Marcion 1 (about 150), who conceives of the Chris-

tian life in a strongly ascetic manner, and regards it as a warfare

against the Demiurge and for the Supreme God revealed through

Jesus, 2 and Marcion's disciple Apelles even treated the Jewish God

1 Cf. Volkmar, Philosophoumena unci Marcion(Theol. Jahrb. Tübingen,

1854). Same author, Das Evangelium Marcion's (Leips. 1852).2 An extremely piquant mythological modification of this thought is found

in the sect of the Ophites, who gave to the Hebraic narrative of the fall the

interpretation, that the serpent which taught man to eat of the tree of knowl-edge in Paradise made a beginning of bringing the revelation of the true Godto man who had fallen under the dominion of the Demiurge, and that after

man had on this account experienced the wrath of the Demiurge, the revela-

tion had appeared victorious in Jesus. For this knowledge which the serpent

desired to teach is the true salvation of man.

Page 279: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 21.] Philosophy of History : Gnostics, Patristics. 259

as Lucifer, who brought carnal sin into the world of sense which

had been formed by the good " Demiurge," the highest angel, so that,

at the petition of the Demiurge, the Supreme God sent the Re-

deemer against him.

5. In contrast with this view we find the doctrine firmly held,

not only by the Recognitions, 1 ascribed to Clement of Rome (which

arose about 150 a.D.), but in the entire orthodox development of

Christian doctrine, that the Supreme God and the creator of the

world, the God of the New and the God of the Old Testaments, are

the same. But a well-planned educative development of the divine

revelation is assumed, and in this the history of salvation, i.e. the

inner history of the world, is sought. Proceeding in accordance

with the suggestions of the Pauline epistles,2 Justin, and especially

Irenseus, took this standpoint. The theory of revelation did not

become complete until it found this elaboration in the philosophy

of history (cf. § 18).

For the anticipations of Christian revelation, that emerge on the

one hand in Jewish prophecy, on the other in Hellenic philosophy,

are regarded from this point of view as pedagogic preparations for

Christianity. And since the redemption of sinful man constitutes,

according to the Christian view, the sole significance and value of

the world's history, and so of all that is real aside from God, the

well-ordered succession of God's acts of revelation appears as the

essential thing in the entire course of the world's events.

In the main, corresponding to the doctrine of revelation, three

stages of this divine, saving activity are distinguished. 3 As divided

theoretically there are, first, the universal-human revelation, given

objectively by the purposiveness of Nature, subjectively through

the rational endowment of the mind ; second, the special revelation

imparted to the Hebrew people through the Mosaic law and the

promises of the prophets ; and third, the complete revelation through

Jesus. Divided according to time, the periods extended from Adamto Moses, from Moses to Christ, from Christ to the end of the world. 4

This triple division was the more natural for ancient Christianity,

the stronger its faith that the closing period of the world's redemp-

1 Edited by Gersdorf (Leips. 1838). Cf. A. Hilgenfeld, Die clementinischen

Recognitionen unci Homihen (Jena, 1848); G. Uhlhoru, Die Homilien unci

Recoijnitionen des CI. B. (Göttingen, 1854).2 Which treat the " law " as the " schoolmaster" unto Christ (7rcu5<x7ü;7ds eis

XpLcrröv); Gal. iii. 24.3 This had been done in part already by the Gnostics, by Basilides at least,

according to Hippolytus.4 The later (heretical) development of eschatology added to these three

periods yet a fourth, by the appearance of the " Paraclete." Cf., e.g., Ter«tullian, be Virg. Vel. 1, p. 884 O.

Page 280: Windelband, History of Philosophy, Vol. 1

260 Hellenistic-Roman Thought : Religious Period. [Pakt IL

tion, which had begun with the appearance of the Saviour, would be

ended in a very short time. The eschatological hopes are an essential

constituent of the early Christian metaphysics ; for the philosophy

}f history which made Jesus the turning-point of the world's history

had, as by no means its slighest support, the expectation that the

Crucified would return again to judge the world, and to complete

the victory of light over darkness. However varied these ideas

become with time and with the disappointment of the first hopes,

however strongly the tendencies of dualism and monism assert

themselves here also, by conceiving of the last Judgment either as

a definite separation of good and evil, or as a complete overcoming

of the latter by the former (d7roK<n-ao-Ta(ns irdvTwv with Origen), and

however much a more material and a more spiritual view of blessed-

ness and unhappiness, of heaven and hell, interplay here also, — in

every case the last Judgment forms the conclusion of the work of

redemption, and so the consummation of the divine plan of salva-

tion.

6. The points of view from which the world's history is regarded

by Christian thinkers are thus indeed exclusively religious ; but the

more general principle of a historical teleology gains recognition

within them. While Greek philosophy had reflected upon the pur-

posiveness of Nature with a depth and an energy which religious

thought could not surpass, the completely new thought rises here

that the course of events in human life also has a purposeful mean-

ing as a whole. The teleology of history becomes raised above

that of Nature, and the former appears as the higher in worth, in

whose service the latter is employed. 1

Such a conception was possible only for a time that from a ripe

result looked back upon the vivid memory of a great development

in the world's history. The universal civilisation of the RomanEmpire found dawning in the self-consciousness of its own inner

life the presentiment of a purpose in that working together of

national destinies through which it had itself come into existence,

and the idea of this mighty process was yielded especially by the

continued tradition of Greek literature embracing a thousand years.

The religiovTS theory of the world, which had developed from this

ancient civilisation, gave to that thought the form that the meaning

of the historical movement was to be sought in the preparations of

God for the salvation of man ; and since the peoples of the ancient

civilisation themselves felt that the time of their efficient working

was complete, it is comprehensible that they believed they saw the

i Cf. Irenseus, Bef. IV. 38, 4, p. 702 f. ,%

Page 281: Windelband, History of Philosophy, Vol. 1

Chap, 2, § 21.] Philosophy of History : Patristics. 261

end of history immediately before them, where the sun of their day

was sinking.

But hand in hand with this idea of a systematically planned unity

in human history goes the thought of a unity of the human race,

exalted above space and time. The consciousness of common civil-

isation, breaking through national boundaries, becomes complete in

the belief in a common revelation and redemption of all men. Inas-

much as the salvation of the whole race is made the import of the

divine plan for the world, it appears that among the provisions of

this plan, the most important is that fellowship (eK/cA^o-ia) to which

all members of the race are called, by sharing in faith the same work

of redemption. The conception of the Church, shaped out from the

life of the Christian community, stands in this connection with the

religious philosophy of history, and accordingly, among its constitu-

tive marks or notes, universality or catholicity is one of the most

important.

7. In this way, man and his destiny becomes the centre of the

universe. This anthropocentric character distinguishes the Christian

view of the world essentially from the Neo-Platonic. The latter,

indeed, assigned a high metaphysical position to the human individ-

ual, whose psychico-spiritual nature it even held to be capable of

deification ; it regarded the purposeful connected whole of Nature

also from the (Stoic) point of view of its usefulness for man,— but

never would Neo-Platonism have consented to declare man, whofor it was a part of the phenomena in which divine efficiency

appears, to be the end of the whole.

Just this, however, is the case in the philosophy of the Fathers.

According to Irenceus, man is the end and aim of creation : it is to

him as a knowing being that God would reveal himself, and for his

sake the rest, the whole of Nature, has been created ; he it is, also,

who by abuse of the freedom granted him, made farther revelation

and redemption necessary ; it is he, therefore, for whose sake all

history also exists. Man as the highest unfolding of psychical life

is, as Gregory of Nyssa teaches, the crown of creation, its master

and king : it is creation's destiny to be contemplated by him, and

taken back into its original spirituality. But with Origen, too, menare just those fallen spirits, who, for punishment and improvement,

have been clothed with the world of sense: Nature exists only on

account of their sin, and it will cease again when the historical

process has attained its end through the return of all spirits to the

( rOOd.

Thus the anthropological movement, which at first forced its way

into Greek science only as a shifting of the interest, as a change in

Page 282: Windelband, History of Philosophy, Vol. 1

262 Hellenistic-Roman Thought : Religious Period. [Part IL

the statement of the problem, developed during the Hellenistic-

Roman period to be more and more the real principle from which

the world was considered, and at last in league with the religious

need it took possession of metaphysics. The human race has gained

the consciousness of the unity of its historical connection and re-

gards the history of its salvation as the measure of all finite things.

What arises and passes away in space and time has its true signifi-

cance only in so far as it is taken up into the relation of man to his

God.

Being and Becoming were the problems of ancient philosophy

at its beginning : the conceptions with which it closes are God and

the human race.

Page 283: Windelband, History of Philosophy, Vol. 1

PART III.

THE PHILOSOPHY OF THE MIDDLE AGES.

Rousselot, Etudes sur la Philosophie du Mayen Age. Paris, 1840-42.

B. Haureau, De la Philosophie Scholastique. Paris, 1850.

B. Haurßau, Histoire de la Philosophie Scholastique. Pans, 1872-80.

A. Stöckl, Geschichte der Philosophie des Mittelalters. Mainz, 1864-66.

When the migration of the peoples broke in devastation over the

Roman Empire, and the latter lacked the political strength to

defend itself against the northern barbarians, scientific civilisation,

also, was in danger of becoming completely crushed out; for the

tribes to whom the sceptre now passed brought still less mind and

understanding for the finely elaborated structures of philosophy

than for the light forms of Grecian art. And, withal, ancient civ-

ilisation was in itself so disintegrated, its vital force was so broken,

that it seemed incapable of taking the rude victors into its school.

Thus the conquests of the Greek spirit would have been given

over to destruction beyond hope of rescue, if in the midst of the

breaking down of the old world, a new spiritual power had not

grown strong, to which the sons of the North bowed, and which,

with firm hand, knew how to rescue for the future the goods of

civilisation, and preserve them during the centuries of subversion.

This power was the Christian Church. What the State could not

do, what art and science could not achieve, religion accomplished.

Inaccessible still for the fine workings of aesthetic imagination and

abstract thought, tha Germans were laid hold of in their deepest

feelings by the preaching of the gospel, which worked upon them

with all the power of its grand simplicity.

Only from this point of religious excitation, therefore, could the

process of the appropriation of ancient science by the peoples of

the Europe of to-day begin ; only at the hand of the Church could

the new world enter the school of the old. The natural conse-

quence, however, of this relation was, that at first only that portion

of the intellectual content of ancient civilisation remained alive

263

Page 284: Windelband, History of Philosophy, Vol. 1

264 Mediaeval Philosophy. [Part III.

which had been taken up into the doctrine of the Christian Church,

and that the teaching authority rigidly excluded all else, and espe-

cially that which was opposed to her. By this means, to be sure,

confusion in the youthful mind of these nations, which would not

have been able to comprehend and elaborate much and many kinds

of material, was wisely guarded against ; but thereby whole worlds

of the intellectual life sank to the depth from which they could

only be drawn forth again long after, by toil and conflict.

The Church had grown to its great task of becoming the educator

of the European nations, first of all, because from the invisible

beginnings of a religious society it had developed with steadily

growing power to a unified organisation, which amid the dissolution

of political life presented itself as the only power that was firm and

sure of itself. And since this organisation was supported by the

thought that the Church was called to become the means of bring-

ing the salvation of redemption to all humanity, the religious edu-

cation of the barbarians was a task prescribed by its own nature.

But the Church was all the more able to take this in hand, since in

her inner life she had proceeded with the same certainty amid

numerous deviating paths, and had attained the goal of a unified

and completed system of doctrine. To this was further added the

especially favourable circumstance, that at the threshold of the newepoch she was presented with the sum-total of her convictions,

worked out into the form of a thorough scientific system by a mind

of the first order, — Augustine.

Augustine was the true teacher of the Middle Ages. Not only

do the threads of Christian and Neo-Platonic thought, the ideas of

Origen and of Plotinus, unite in his philosophy, but he also concen-

trated the entire thought of his time with creative energy about the

need of salvation and the fulfilment of this need by the church

communit}^. His doctrine is the philosophy of the Christian Church.

Herewith was given, in pregnant unity, the system which became

the basis of the scientific training of the European peoples, and in

this form the Romanic and Germanic peoples entered upon the

inheritance of the Greeks.

But for this reason the Middle Ages retraced in the reverse direc-

tion the path which the Greeks had gone over in their relations to

science. In antiquity science had arisen from the pure aesthetic joy

in knowledge itself, and had only gradually entered into the service

of practical need, of ethical tasks, and of religious longings. The

Middle Ages begins with the conscious subordination of knowledge

to the great ends of faith ; it sees in science at the beginning only

the task of the intellect to make clear to itself and express in

Page 285: Windelband, History of Philosophy, Vol. 1

Mediaeval Philosophy. 265

abstract thought that which it possesses surely and unassailably

in feeling and conviction. But in the midst of this work the joy

?n knowledge itself wakes anew, at first timorously and uncertainly,

then with ever-increasing force and self-certainty ; it unfolds itself

at first scholastically, in fields which seem to lie far distant from

faith's unassailable sphere of ideas, and at the end breaks through

victoriously when science begins to define her limits as against

faith, philosophy hers as against theology, and to assume a con-

scious independent position.

The education of the European peoples, which the history of the

philosophy of the Middle Ages sets forth, has then for its starting-

point the Church doctrine, and for its goal the development of

the scientific spirit. The intellectual civilisation of antiquity is

brought to modern peoples in the religious form which it assumed

at its close, and develops in them gradually the maturity for prop-

erly scientific work.

Under such conditions it is easy to understand that the history

of this education awakens psychological interest and an interest

connected with the history of civilisation, rather than presents newand independent fruits of philosophical insight. In the appropria-

tion of the presented material the peculiar personality of the

disciple may assert itself here and there ; the problems and con-

ceptions of ancient philosophy may, therefore, find many fine trans-

formations when thus taken up into the spirit of the new peoples,

and in forging out the new Latin terminology in the Middle Ages

acuteness and depth often contend emulously with pedantry and

insipidity ; but in its fundamental philosophical thoughts, mediaeval

philosophy remains enclosed within the system of conceptions of

the Greek and the Hellenistic-Roman philosophy, — not only as

regards its problems, but also as regards their solutions. Highly

as we must estimate the worth of its labours for the intellectual

education of European peoples, its highest achievements remain in

the last instance just brilliant productions of scholars or disciples,

not of masters, — productions in which only the eye of the most

refined detailed investigation can discover the gently germinating

beginnings of a new thought, but which show themselves to be, on

the whole, an appropriation of the world of thought of the depart-

ing antiquity. Mediaeval philosophy is, in its entire spirit, solely

the continuation of the Hellenistic-Roman, and the essential dis-

tinction between the two is that what in the first centuries of our

era had been coming into existence amid struggles was, for the

Middle Ages, given and regarded as something in the main complete

and definitive.

Page 286: Windelband, History of Philosophy, Vol. 1

266 Mediaeval Philosophy. [Part III.

This period, in which the humanity of to-day was at school,

lasted a full thousand years, and as if in systematically planned

pedagogic steps its education proceeds toward science by the suc-

cessive addition of ancient material of culture. Out of the antith-

eses which appear in this material grow the problems of philosophy,

and the ancient conceptions taken up and amplified give the form

to the scientific theories of the world prevalent in the Middle

Ages.

An original discord exists in this tradition between Neo-Platonism

and the Church doctrine defended by Augustine, — a discord which

indeed was not equally strong at all points, since Augustine in very

essential points had remained under the control of Neo-Platonism,

and yet a discord which amounted to an opposition with reference

to the fundamental character of the relation of philosophy to faith.

The system of Augustine is concentrated about the conception of

the Church ; for it philosophy has as its main task to present the

Church doctrine as a scientific system, to establish and develop it

:

in so far as it prosecutes this task mediaeval philosophy is the

science of the schools, Scholasticism. The Neo-Platonic tendency,

on the contrary, takes the direction of guiding the individual,

through knowledge, to blessed oneness of life with the deity : in so

far as the science of the Middle Ages sets itself this end it is Mysti-

cism.

Scholasticism and Mysticism accordingly supplement each other

without being reciprocally exclusive. As the intuition of the Mystics

may become a part of the Scholastic system, so the proclamation of

the Mystics may presuppose the system of the Scholastics as its

background. Throughout the Middle Ages, therefore, Mysticism is

more in danger than Scholasticism of becoming heterodox ; but it

would be erroneous to see in this an essential mark for distinguish-

ing between the two. Scholasticism is, no doubt, m the main

entirely orthodox ; but not only do the theories of the Scholastics

diverge widely in the treatment of dogmas which are still in

the process of formulation, but many of the Scholastics, even in

the scientific investigation of the doctrines which were given, pro-

ceeded to completely heterodox theories, the expression of which

brought them into more or less severe conflicts without and within.

As regards Mysticism, the Neo-Platonic tradition often forms the

theoretical background of the secret or open opposition offered to

the monopolising of the religious life on the part of the Church; 1

1 Cf. H. Reuter, Geschichte der religiösen Aufklärung im Mittelalter, 2 vols.

(Berlin, 1875-77). Cf. also H. v. Eicken, Geschichte der mittelalterlichen Welt-

anschauung (Stuttgart, 1888).

Page 287: Windelband, History of Philosophy, Vol. 1

Mediaeval Philosophy. 267

but we meet on the other hand enthusiastic Mystics who feel them-

selves called to take the true faith into their protection against the

excesses of Scholastic science.

It appears thus to be inappropriate to give to the philosophy of

the Middle Ages the general name of " Scholasticism." It might

rather prove, as the result of a more exact estimate, that in the

maintenance of scientific tradition as well as in the slow adaptation

and transformation of those philosophical doctrines which were

effective for the after time, a part belongs to Mysticism which is

at least as great as the part played by Scholasticism, and that on the

other hand a sharp separation of the two currents is not practicable

in the case of a great number of the most prominent philosophic

thinkers of the Middle Ages.

Finally, it must be added that even when we put together Scholas-

ticism and Mysticism, we have in nowise exhausted the character-

istics of mediaeval philosophy. While the nature of both these

tendencies is fixed by their relation to the religious presuppositions

of thought,— in the one case the established doctrine of the Church,

in the other personal piety, — there runs along side by side with

these, especially in the later centuries of the Middle Ages though

noticeable still earlier, a secular side-current which brings in an in-

creasing degree the rich results of Greek and Koman experience of

the world, to science building itself anew. Here, too, at the outset

the effort prevails to introduce organically into the Scholastic

system this extensive material and the forms of thought which are

dominant in it ; but the more this part of the sphere of thought

develops into an independent significance, the more the entire lines

of the scientific consideration of the world become shifted, and

while the reflective interpretation and rationalisation of the relig-

ious feeling becomes insulated within itself, philosophical knowl-

edge begins to mark off anew for itself the province of purely

theoretical investigation.

From this multiplicity of variously interwoven threads of tradi-

tion with which ancient science weaves its fabric on into the Middle

Ages, we can understand the wealth of colour in which the philosophy

of this thousand years spreads out before historical research. In

the frequent exchange of friendly and hostile contact, these elements

of a tradition changing in compass and content from century to

century play back and forth to form ever new pictures ; a surprising

fineness in the transitions and shadings becomes developed as these

elements are woven together, and thus there is developed also a

wealth of life in the work of thought, which manifests itself in a

considerable number of interesting personalities, in an astonishing

Page 288: Windelband, History of Philosophy, Vol. 1

268 Mediaeval Philosophy. [Pari' IIL

amount of literary production, and in a passionate agitation of scien-

tific controversies.

Such living variety in form has as yet by no means everywhere

received full justice at the hands of literary-historical research, 1 but

the main lines of this development lie before us clearly and dis-

tinctly enough for the history of philosophic principles, which

nevertheless finds but a meagre field in this period for the reasons

already adduced. We must, indeed, be on our guard against aiming to

reduce the complex movement of this process to formulas that are

all too simple, and against overlooking the multitude of positive

and negative relations that have come and gone in shifting forms

between the elements of ancient tradition which found their en-

trance in the course of centuries by irregular intervals into mediaeval

thought.

In general, the course of science among the European peoples of

the Middle Ages proceeded along the following lines.

The profound doctrine of Augustine had its first efficiency, not in

the direction of its philosophical significance, but as an authoritative

presentation of the doctrine of the Church. Side by side with this

a Neo-Platonic Mysticism maintained itself, and scientific schooling

was limited to unimportant compendiums, and to fragments of the

Aristotelian logic. Nevertheless, a logico-metaphysical problem of

great importance developed from the elaboration of the logic,

and about this problem arose a highly vigorous movement of

thought, which, however, threatened to degenerate into barren for-

malism in consequence of the lack in knowledge to form the content

of thought. In contrast with this the Augustinian psychology

began gradually to assert its mighty force ; and at the same time the

first effects of contact with Arabian science disclosed themselves, a

science to which the West owed, primarily at least, a certain stimulus

toward employment with realities, and further a complete widening

1 The grounds for this lie, certainly in part, in the but gradually vanishingprejudices which long stood in the way of a just appreciation of the MiddleAges ; but in no less a degree they lie also in this literature itself. The circum-stantial and yet for the most part sterile prolixity of the investigations, theschematic uniformity of the methods, the constant repetition and turning of

the arguments, the lavish expenditure of acuteness upon artificial and sometimesabsolutely silly questions, the uninteresting witticisms of the schools,— all these

are features which perhaps belong inevitably to the process of learning, appro-priating, and practising, which mediaeval philosophy sets forth, but they bringwith them the consequence that in the study of this part of the history of phi-

losophy the mass of the material, and the toil involved in its elaboration, standin an unfavourable relation to the real results. So it has come about that just

those investigators who have gone deeply, with industry and perseverance, into

mediaeval philosophy have often not refrained from a harsh expression of ill-

humour as to the object of thefr research.

Page 289: Windelband, History of Philosophy, Vol. 1

Mediaeval Philosophy. 269

and transformation of its horizon. This development was in the

main attached to the acquaintance gained by such by-ways with the

entire system of Aristotle, and the immediate consequence of this

acquaintance was that the structure of Church doctrine was pro-

jected in the grandest style and carefully wrought out in all its

parts with the help of his fundamental metaphysical conceptions.

Meanwhile Aristotelianism had been accepted from the Arabians

(and Jews) not only in their Latin translation, but also with their

commentaries, and in their interpretation which was under strong

Neo-Platonic influence ; and while by this means the Neo-Platonic

elements in previous tradition, even in the Augustinian form, found

vigorous confirmation in various directions, the specific elements of

the Augustinian metaphysics were forced into sharper and more

energetic expression, in violent reaction against the Neo-Platonic

tendency. Thus while both sides lean upon Aristotelianism, a cleft

in scientific thought is produced, which finds its expression in the

separation of theology and philosophy. This cleft became widened

by a new and not less complicated movement. Empirical research

in medicine and natural science had also made its way from the

East, hand in hand with Aristotelianism ; it began now to rise also

among the European peoples ; it conquered the domain of p>sychology

not without assistance from the Augustinian current, and favoured

the development of the Aristotelian logic in a direction which led

far from the churchly Aristotelian metaphysics. And while thus

the interwoven threads of tradition were separating on all sides, the

fine filaments of new beginnings were already finding their way into

this loosening web.

With such various relations of mutual support or retardation,

and with such numerous changes of front, the thoughts of ancient

philosophy move through the Middle Ages ; but the most important

and decisive turn was doubtless the reception ofAristotelianism, which

became complete about the year 1200. This divides the whole

field naturally into two sections which in their philosophical import

are so related that the interests and the problems, the antitheses

and the movements, of the first period are repeated in broader, and

at the same time deeper, form in the second. The relation of these

two divisions, therefore, cannot be generally designated in this case

by differences in the subject matter.

Page 290: Windelband, History of Philosophy, Vol. 1

CHAPTER I. FIRST PERIOD.

(Until about 1200.)

W. Kaulich, Geschichte der sclwlastichen Philosophie, I. Theil. Prague, 1863.

The line of thought in which mediaeval philosophy essentially

moved, and in which it continued the principles of the philosophy

of antiquity, was prescribed for it by the doctrine of Augustine.

He had moved the principle of internality (Innerlichkeit), which

had been preparing in the whole closing development of ancient

science, for the first time into the controlling central position of

philosophic thought, and the position to which he is entitled in

the history of philosophy is that of the beginner of a new line of

development. For the bringing together of all lines of the Patristic

as well as the Hellenistic philosophy of his time, which he com-

pletely accomplished, was possible only as these were consciously

united in that new thought which was itself to become the germ of

the philosophy of the future. But only of a more distant future :

his philosophical originality passed over his contemporaries and the

immediately following centuries without effect. Within the circuit

of the old civilisation the creative power of thought had become

extinguished, and the new peoples could only gradually grow into

scientific work.

In the cloister and court schools which formed the seats of this

newly beginning civilisation, permission for instruction in dialectic

by the side of the arts most necessary for the training of the clergy

had to be conquered step by step. For this elementary logical

instruction they possessed in the first centuries of the Middle Ages

only the two least important treatises of the Aristotelian Organon,

De Categoriis and De Interpretatione, in a Latin translation with

the introduction of Porphyry, and a number of commentaries of

the Neo-Platonic time, in particular those of Boethius. For the

material of knowledge (of the Quadrivium) they used the com-

pendiums of departing antiquity, which had been prepared by

Marcianus Capella, Cassiodorus, and Isidorus of Sevilla. Of the

270

Page 291: Windelband, History of Philosophy, Vol. 1

Chap. 1.] First Period. 271

great original works of ancient philosophy, only tne Platonic

Timceus in the translation of Chalcidius was known.

Under these circumstances, scientific activity in the schools was

mainly directed toward learning and practising the schematism of

formal logic, and the treatment even of the material parts of knowl-

edge, in particular of religious dogma which was indeed regarded

as something essentially complete and in its contents unassailable,

took the direction of elaborating and setting forth what was given

and handed down b}^ tradition, in the forms and according to the

rules of the Aristotelian-Stoic logic. In this process the main em-

phasis must necessarily fall upon formal arrangement, upon the

formation and division of class-concepts, upon correct syllogistic

conclusions. Already in the Orient the ancient school logic had

been put into the service of a rigidly articulated development of

Church doctrine by John Damascenus, and now this took place in

the schools of the West also.

Meanwhile this pursuit, which had its basis in the conditions of

the tradition, had not only the didactic value of a mental exercise

in the appropriation of material, but also the consequence that the

beginnings of independent reflection necessarily took the direction

of an inquiry as to the significance of logical relations, and so we

find emerging early in the Western literature, investigations as to

the relation of the conception on the one hand to the word, and on

the other to the thing.

The problem thus formed became strengthened by a peculiar com-

plication. By the side of the Church doctrine there persisted, half

tolerated and half condemned, a mystical transmission of Chris-

tianity in Neo-Platonic form. It went back to writings which had

arisen in the fifth century, but which were ascribed to Dionysius

the Areopagite, and it gained wider extension when these writings

were translated in the ninth century by John Scotus Erigena, and

made the basis of his own doctrine. In this doctrine, however,

a main point was that identification of the different grades of ab-

straction with the stages of metaphysical reality, which had been

already propounded in the older Platonism and in Neo-Platonism

(cf. § 20, 8).

In consequence of these incitements the question as to the meta-

physical significance of logical genera became, during the next centuries,

the centre of philosophic thought. About this were grouped the

other logical and metaphysical problems, and the answer given to

this question decided the party position of individual thinkers.

Amid the great variety of decisions given in this controversy over

universale, three tendencies are prominent: Realism, which main-

Page 292: Windelband, History of Philosophy, Vol. 1

272 Mediceval Philosophy. [Part III.

tains the independent existence of genera and species, is the doctrine

of Anselm of Canterbury, of William of Champeaux, and of the

Platonists proper, among whom Bernard of Chartres is prominent

;

Nominalism, which sees in universals only designations or terms

which apply commonly, is defended in this period principally by

Roscellinus; finally a mediating theory, which has been called

Conceptualism or Sermonism, is attached principally to the name of

Abelard.

These conflicts came to an issue orincipally in the endless dispu-

tations at the Paris University, which for this period and on into

the following period formed the centre of scientific life in Europe;

and these battles, conducted with all the arts of dialectical dexterity,

exercised upon this age a fascinating power like that which the

disputes of the Sophists and Socratic circles had once exercised

upon the Greeks. Here as there the unreflective life of the popular

consciousness was awakened to thought, and here as there wider

circles were seized by a feverish thirst for knowledge, and by a pas-

sionate desire to take part in such hitherto unwonted intellectual

games. Far beyond the narrow circles of the clergy, who had pre-

viously been the transmitters of scientific tradition, the impulse

toward knowledge, thus awakened, forced its way to the surface.

But this excessive vigour in dialectical development found at the

same time manifold opposition. In fact, it hid within itself a seri-

ous danger. This brilliant performance, in which abstract thought

proved its power, lacked all basis of real knowledge. With its dis-

tinctions and conclusions it was carrying on to a certain extent a

juggler's game in the open air, which indeed set the formal mental

powers into beneficial motion, but which, in spite of all its turns and

windings, could lead to no material knowledge. Hence, from intelli-

gent men like Gerbert, who had received information from the empir-

ical studies of the Arabians, went out the admonition to abandon

the formalism of the schools and turn to the careful examination

of Nature and to the tasks of practical civilisation.

But while such a call still echoed mainly unheard, dialectic met a

more forcible resistance in the piety of faith and in the power of the

Church. The result was inevitable that the logical working over of

the metaphysics of the Church's faith, and the consequences which

were developed in the strife about universals, — at first without any

reference to their religious bearing,— should come into contradiction

with the dogma of the Church ; and the more this was repeated, the

more dialectic appeared, not only superfluous for the simply pious

mind, but also dangerous to the interests of the Church. In this

spirit it was attacked, sometimes with extreme violence, by the

Page 293: Windelband, History of Philosophy, Vol. 1

Chap. 1.] First Period. 273

orthodox Mystics, among whom the most combative was Bernard of

Clairvaux, while the Victorines turned back from the excesses of

dialectical arrogance to the study of Augustine, and sought to bring

out the rich treasure of inner experience which his writings con-

tained, by transferring the fundamental thoughts of his psychology

from the metaphysical to the empirical sphere.

Aurelius Augustinus (354-430) , born at Thagaste in Numidia, and educatedfor a jurist there and also in Madaura in Carthage, passed through in his youthalmost all phases of the scientific and religious movement of his time. Hesought at first in Manichseism religious relief for his burning doubts, then fell

into the Academic Scepticism which he had early absorbed from Cicero, passedover from this gradually to the Neo-Platonic doctrine, and was at last won byAmbrose, Bishop of Milan, for Christianity, whose philosopher he was to become.

As priest, and later as bishop at Hippo Regius, he was unwearied in practical

and literary activity for the unity of the Christian Church and doctrine; his

doctrinal system was developed especially in the Donatist and Pelagian contro-

versies. Among his works (in Migne's collection, 16 vols., Paris, 1835 ff. [tr.

ed. by Dods, 15 vols., Edin. 1871-77 ; also in Schaffs lib., Nicene and Post-

Nicene Fathers, Vols. 1-8, Buffalo, 1886-88] ) those of chief importance for

philosophy are his autobiographical Confessions, and further Contra Academi-cos, De Beata Vita, De Online, De Quantitate Animce, De Libero Arbitrio, DeTrinitate, Soliloquia, De Immortalitate Animce, De Civitate Dei.— Cf. C. Binde-mann, Der. hlg. A. (3 Bde. 1844-1869).— Fr. Böhringer, Kirchengeschichte in

Biographien, XL Bd. in 2 Thl. (Stuttgart, 1877-78).— A. Dorner, A. (Berlin,

1873). — W. Dilthey, Einleitung in die Geistesioissenschaften, I. (Leips. 1883),

pp. 322 ff. — J. Storz, Die Philos. des hlg. A. (Freiburg, 1892).

The EtsaYw7T? eis ras Kar-qyoplas of Porphyry (ed. by Busse, Berlin, 1887), in

its translation by Boethius, gave the external occasion for the controversy over

universals. Boethius (470-525), aside from this, exercised an influence uponthe early Middle Ages by his translations and commentaries upon the twoAristotelian treatises, and upon a number of Cicero's writings. In addition to

his books there were still others which circulated under the name of Augustine.

Cf. Prantl, Gesch. d. Log. im Abendl., IL, and A. Jourdain, Becherches critiques

sur Vage et Vorigine des traductioyis latines d' Aristotle (Paris, 2 ed., 1843).

Among the scientific encyclopedias of departing antiquity, Marcianus Capella

(from Carthage, the middle of the fifth century), in his Satyricon (ed. byEyssenhardt, Leips. 1866), after his whimsical introduction De Xuptiis Mercurii

et Philologice, treats the seven liberal arts, of which, as is well known, in the

activity of the schools grammar, rhetoric, and dialectic formed the Trivium r

arithmetic, geometry, astronomy, and music, including poetics, the QuadriviunuA valuable commentary on Capella was written later by Scotus Erigena (ed. byB. Haureau, Paris, 1861).— The Institutions Divinarum et Sacidarium Lec-

tionum and De Artibus ac Disciplinis Litterarum Liberalium of the Senator Cas-

siodorus (480-570, Works, Paris, 1588), and the Originum sive Etymologiarum,,LibriXX. (in Migne) of Isidorus Hispalensis (died 036) are already completelyupon theological ground. John Damascenus (about 700) in his II77777 yvüaews

(Works, Venice, 1748) gave the classical example for the employment of the

ancient school logic in the service of systematising the Church doctrines.

While the storms of the national migrations were blustering upon the conti-

nent, scientific study had fled to the British Isles, in particular to Ireland, andlater flourished to a certain extent in the school at York under the VenerableBede. From here learned education was won back to the continent throughAlcuin, upon the inducement of Charles the Great ; beside the episcopal and the

cloister schools arose the palatinal school, whose seat was fixed by Charles the

Bald at Paris. The most important cloister schools were those of Fulda andTours. At the former worked Rabanus (Rhaban) Maurus (of Mainz, 770-856

;

De Universo, Libri XXII.), and Eric (Ileiricus) of Auxerre; from it went out,

at the end of the ninth century, Remigius of Auxerre and the probable author

Page 294: Windelband, History of Philosophy, Vol. 1

274 Mediceval Philosophy. [Part III

of the commentary Super Porphyrium (printed in Cousin's Oicvrages Inedits

d'Abelard, Paris, 1836). In Tours Alcuin was followed by the Abbot Prede-gisus, whose letter, De Nihilo et Tenebris, is preserved (in Migne, Vol. 105).

Later the cloister at St. Gall (Notker Labeo, died 1022) formed a principal seat

of scientific tradition.

Cf. also for the literary relations, the Histoire Litteraire de la France.

The writings ascribed to the Areopagite (cf. Acts of the Apostles, 17 : 34),among which those of chief importance are irepl ßvartKijs öeoXoytas and irepl t^s

tepapx'as ovpavlov (in Migne ; German by Engelhardt, Sulzbach, 1823), show the

same mixture of Christian and Neo-Platonic philosophy which appeared fre-

quently in the Orient (the result of Origen's influence) and in an especially

characteristic form in the Bishop Synesius (about 400 ; cf. R. Volkmann, S. vonGyrene, Berlin, 1869). The above-named writings of the Pseudo-Dionysius,

which probably arose in the fifth century, are first mentioned, 532, and their

genuineness is there contested; nevertheless, this was defended by MaximusConfessor (580-662 ; De Varüs Difficüioribus Locis Patrum Dionysii et Gregorii,

ed. Oehler, Halle, 1857).

In connection with this Mysticism develops the first important scientific

personality of the Middle Ages, John Scotus Erigena (sometimes Jerugena,from Ireland, about 810-880), of whose life it is certainly known that he wascalled by Charles the Bald to the court school at Paris, and was for a timeactive there. He translated the writings of the Areopagite, wrote against

Gottschalk the treatise De Praedestinatione, and put his own theories into his

main work, De Divisione Natura; (German by Noack, Leips. 1870-76). Theworks form Vol. 122 in Migne's collection. Cf. J. Huber, J. S. E. (Munich, 1861).

Anselm of Canterbury (1033-1109) came from Aosta, was active for a long

time in the Norman cloister at Bee, and was called to become Archbishop of

Canterbury in 1093. Of his works (Migne, Vol. 155) the most important for

philosophy besides the treatise Cur Deus Homo? are the Monologium and the

Proslogium. The two latter are edited by C. Haas (Tübingen, 1863), togetherwith the refutation of a monk, Gaunilo (in the cloister Marmoutier near Tours),Liber pro Insipiente, and the reply of Anselm. Cf. Ch. R§musat, A. de C,tableau de la vie monastique et de la lutte du pouvoir spirituel avec le pouvoirtemporel au ll me siecle (2d ed., Paris, 1868).

William of Champeaux (died 1121 as Bishop of Chalons-sur-Marne) was ateacher who was much heard at the cathedral school in Paris, and establishedstudies there in the Augustinian cloister at St. Victor. We are chiefly informedas to his philosophical views by his opponent Abelard ; his logical treatise is lost.

Cf. E. Michaud, G. de Oh. et les ecoles de Paris au 12me siecle (Paris, 1868).

The Platonism of the earlier Middle Ages attached itself essentially to theTimoeus, and under the influence of the Neo-Platonic interpretation gave to thedoctrine of Ideas a form which did not completely correspond to the original

sense. The most important figure in this line is Bernard of Chartres (in the

first half of the twelfth century). His work De Mundi Universitate sive Mega-cosmus et Microcosmus has been edited by C. S. Barach (Innsbruck, 1876).

William of Conches (Magna de Naturis Philosophia ; Dragmaticon Philoso-phke) and Walter of Montagne are regarded as his disciples. Adelard of

Bath also wrote in the same spirit (De Eodem et Diverso ; Questiones Naturales).Roscellinus of Armorica in Brittany came forward as teacher at various

places, especially at Locmenach where Abelard was his hearer, and wasobliged to retract his opinions at the Council at Soissons. Of his own writingsonly a letter to Abelard is extant (printed in the Abhandl. der bair. Akad., 1851) ;

the sources for his doctrine are Anselm, Abelard, John of Salisbury.

Abelard (Abeillard) , the most impressive and energetic personality amongthe thinkers of this period, was born 1079 at Pallet, in the county of Nantes,and was a pupil of William of Champeaux and of Roscellinus. His own activity

as a teacher was developed at Melun and Corbeil, and most successfully in

Paris at the cathedral school, and at the logical school St. Genevieve. Themisfortune into which his well-known relationship to Heloise plunged him, andthe conflicts into which his teaching brought him with the Church authority,

chiefly at the instigation of his unwearied prosecutor, Bernard of Clairvaux

Page 295: Windelband, History of Philosophy, Vol. 1

Chap. 1.] First Period. 275

(Synods at Soissons 1121, and Sens 1141), did not allow the restless man to

attain complete clearness in his mind, and impelled him to seek resting-places

in various cloisters : he died 1142 in St. Marcel, near Chälons-sur-Saöne. Cf. his

Historia Calamitatum Meartim, and his correspondence with Heloise (M. Car-

riere, A. u. H., 2d ed., Giessen, 1853). His works have been edited by V. Cousinin two volumes (Paris, 1849-59). Among these the most important are hi?-

Dialectic, Introductio in Theologium, Theologia Christiana, Dialogus intet

Philosophum, Christianum et Judoeum, the treatise Sic et Non, and the ethical

treatise Scito Te Ipsum. Cf. Ch. d. Remusat, Ahelard (2 vols., Paris, 1845).

A number of anonymous treatises (published by V. Cousin) occupy a position

allied to that of Abelard. Of this description are a commentary on De Interpre-

tations, De Intellectibus, and De Generibus et Speciebus (the latter is possibly

from Joscellinus, a Bishop of Soissons who died 1151). Related to Abelard is

also the philosophico-theological position of Gilbert de la Porree (Gilbertus

Porretanus, died 1154 as Bishop of Poitiers), who taught in Chartres and Paris,

and was drawn into the prosecution of Abelard by Bernard of Clairvaux.

Besides a commentary on the De Trinitate and De Dudbus Nahiris in Christo

of Pseudo-Boethius, he wrote the De sex Principiis, which was much com-mented upon later.

The consequences of the " dialectic" that were objectionable for the Churchshowed themselves at an early date especially with Berengar of Tours (999-

1088), whose doctrine of the Sacrament was combated by Lanfranc (1005-

1089, Anselm's predecessor at Bee and Canterbury). The latter is probably

the author of the treatise formerly ascribed to Anselm and printed among his

works, Elucidarium sive Dialogus Summam Totius Theologia} Complectens.

In this compendium the effort first appears to give the whole compass of whathad been established by the Church, in the form of a logically arranged text-

book, putting aside dialectical innovations. From this proceeded later the

works of the Summists [so called from their writings which took the form of

a "Sum" of theology], among whom the most important is Peter Lombard(died 1164 as Bishop of Paris). His Libri IV. Sententiarum form Vol. 192 in

Migne. Among the earlier we may perhaps mention Kobert Pulleyn (RobertusPullus, died 1150) ; among the later, Peter of Poitiers (died 1205) and AlanusRyssel ("a& insulis'1 ''

; died 1203). Cf. on him Baumgartner (Münster, 1896).

Gerbert (died 1003 as Pope Sylvester II.) has the merit of having pointed

out energetically the necessity of the study of mathematics and natural science.

He became acquainted with the work of the Arabians while in Spain and Italy,

and acquired an amount of knowledge that made him an object of amazementand suspicion to his contemporaries. Cf. K. Werner, G-. von Auriliac, die

Kirche und Wissenschaft seiner Zeit (2d ed., Vienna, 1881). Like him his

disciple, Fulbert (died 1029 as Bishop of Chartres), called men back fromdialectic to simple piety, and in the same spirit Hildebert of Lavardin wasactive (1057-1133, Bishop of Tours).

The same thing was done upon a large scale by the orthodox Mysticism of

the twelfth century. As its most zealous supporter we are met by Bernard of

Clairvaux (1091-1 153) . Among his writings those prominent are De ContemptuMundi, and De Gradibus Humilitatis (ed. by Mabillon, last ed., Paris, 1839 f.).

Cf. Neander, Der heilige B. und seine Zeit (3d ed., 1865) ;Morison, Life and

Times of St. B. (Lond.' 1868) ;[R. S. Storrs, B. of C. (N.Y. 1892)].

Mysticism became scientifically fruitful among the Victorines, the conduc-

tors of the cloister school of St. Victor, in Paris. The most important was Hugoof St. Victor (born 1096 as Count of Blankenburg in the Harz, died 1141).

Among his works (in Migne, Vols. 175-177) the most important is De Sacra-

mentis Fidei Christiance; for the psychology of Mysticism the mast importantworks are the Solüoquium de Arrha Animue, De Area Noe and De Vanitate

Mundi, and besides these the encyclopedic work Eruditio Didascalica. — Cf. A.

Liebner, H. v. St. V. und die theologischen Bichtungen seiner Zeit (Leips. 1836).

His pupil, Richard of St. Victor (a Scot, died 1173), wrote De Statu, DeEruditione Hominis Interioris, De Preparatione Animi ad Contemplationem,<md De Gratia Contemplationis. His works form Vol. 194 in Migne. Cf.

W. A. Kaulich, Die Lehren des H. und B. von St. V. (in the Ahhandl. der

Böhm. Ges. der Wiss., 1863 f.). His successor, Walter of St. Vietor. distin-

Page 296: Windelband, History of Philosophy, Vol. 1

276 Mediaeval Philosophy : First Period. [Part III.

guished himself in a less scientific polemic against the heretical dialectic (InQuattuor Labyrinthos Francice).At the close of this period appear the beginnings of a Humanist reaction

against the one-sidedness of the work of the schools, in John of Salisbury(Johannes Saresberiensis, died 1180 as Bishop of Chartres), whose writings Poli-craticus and Metalogicus (Migne, Vol. 199) form a valuable source for thescientific life of the time. Cf. C. Schaarschmidt, J. S. nach Leben und Studien,Schriften und Philosophie (Leips. 1862).

§ 22. The Metaphysics of Inner Experience.

The philosophy of the great Church teacher Augustine is not

presented in any of his works as a complete system ; rather, it

develops incidentally in all his literary activity in connection with

the treatment of various subjects, for the most part theological.

But from this work as a whole we receive the peculiar impression

that these rich masses of thought are in motion in two different

directions, and are held together only by the powerful personality

of the man. As theologian Augustine throughout all his investi-

gations keeps the conception of the Church in mind, as criterion ; as

philosopher he makes all his ideas centre about the principle of the

absolute and immediate certainty (Selbstgeivissheit) of consciousness.

By their double relation to these two fixed postulates, all questions

come into active flux. Augustine's world of thought is like an

elliptic system which is constructed by motion about two centres,

and this, its inner duality, is frequently that of contradiction. 1

It becomes the task of the history of philosophy to separate from

this complicated system those ideas by which Augustine far tran-

scended his time and likewise the immediately following centuries,

rmd became one of the founders of modern thought. All these ideas,

however, have their ultimate ground and inner union in the prin-

ciple of the immediate certainty of inner experience {selbstgewissen

Innerlichkeit), which Augustine first expressed with complete clear-

ness, and formulated and used as the starting-point of philosophy.

Under the influence of the ethical and religious interest, metaphys-

ical interest had become gradually and almost imperceptibly shifted

from the sphere of the outer to that of the inner life. Psychical

conceptions had taken the place of physical, as the fundamental

factors in the conception of the world. It was reserved for Augus-

tine to bring into full and conscious use, this, which had already

become an accomplished fact in Origen and Plotinus. 2

1 It is unmistakable that Augustine himself in the course of his developmenttransferred the emphasis of his personality more and more from the philosophi-

cal to the Church centre. This comes forward with especial distinctness in his

backward look over his own literary activity, the Betractationes.- Aug. De Ver. Bel. 39, 72. Noll foras ire; in te ipsum redi: in interioke

nomine habitat Veritas.

Page 297: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 22.] Metaphysics of Inner Experience : Augustine. 'Ill

This tendency toward inner experience even constitutes his pecu-

liar literary quality. Augustine is a virtuoso in self-observation

and self-analysis ; he has a mastery in the portrayal of psychical

states, which is as admirable as is his ability to analyse these in

reflection and lay bare the deepest elements of feeling and impulse.

Just for this reason it is from this source almost exclusively that

he draws the views with which his metaphysics seeks to compre-

hend the universe. So there begins, as over against the Greek

philosophy, a new course of development, which indeed, during

the Middle Ages, made but little progress beyond what was achieved

by Augustine in his first cast, and the full development of which is

not to be found until the modern period.

1. This makes its appearance clearly already in Augustine's

doctrine of the starting-point of philosophical knowledge. In cor-

respondence with the course of his personal development he seeks

the way to certainty through doubt, and in this process, sceptical

theories themselves must break the path. At first, to be sure, with

the indomitable thirst of his ardent nature for happiness, he

strikes down doubt by the Socratic postulate that the possession of

truth (without the presupposition of which there is also no proba-

bility) is requisite for happiness, and therefore is to be regarded as

attainable : but with greater emphasis he shows that even the

sceptic who denies the external reality of the content of perception,

or at least leaves it undecided, can yet not involve in doubt the

internal existence of the sensation as such. But instead of con-

tenting himself with the relativistic or positivistic interpretations

of this fact, Augustine presses forward just from this basis to victo-

rious certainty. He points out that together with the sensatior

there is given not only its content, which is liable to doubt in one

direction or another, but also the reality of the perceiving subject,

and this certainty which consciousness has in itself follows first of

all from the very act of doubt. In that I doubt, or since I doubt,

he says, I know that I, the doubter, am : and thus, just this doubt

contains within itself the valuable truth of the reality of the con-

scious being. Even if I should err in all else, I cannot err in this;

for in order to err I must exist. 1

This fundamental certainty extends equally to all states of con-

1 Augustine attributed fundamental importance to this line of argument,which he frequently worked out (De Beata Vita, 7 ; Solu. II. 1 ff.

; D<> Vir.

Eel. 72 f. ; De Trin. X. 14, etc.). That it, however, was not completelyunknown to Greek literature also is proved by the passage (III. 6 f.) of thecompilation current under the name of "Metaphysics of Herennios." Thesource of this passage has not as yet been discovered, but is probably late Stoic.Cf. on this E. Heitz in Sitz.-Bcr. der Berl. Ak. d. W., 1*89, pp. 1167 ff.

Page 298: Windelband, History of Philosophy, Vol. 1

278 Mediaeval Philosophy : First Period. [Part III

sciousness {cogitate), and Augustine sought to show that all the

various kinds of these states are already included in the act of

doubt. He who doubts knows not only that he lives, but also that

he remembers, that he knows, and that he wills : for the grounds

of his doubt rest upon his former ideas ; in estimating the momenta

of the doubt are developed thought, knowledge, and judgment; and

the motive of his doubt is only this, that he is striving after truth.

Without particularly reflecting upon this, or drawing farther con-

elusions from it, Augustine proves in this example his deep insight

into the psychical life, since he does not regard the different kinds

of psychical activity as separate spheres, but as the aspects of one

and the same act, inseparably united with one another. The soul

is for him — and by this he rises far above Aristotle, and also above

the Neo-Platonists — the living whole of personality, whose life is

a unity, and which, by its self-consciousness, is certain of its ownreality as the surest truth.

2. But from this first certainty Augustine's doctrine at once

leads farther, and it is not only his religious conviction, but also

a deep epistemological reflection, that makes him regard the idea

of God as immediately involved in the certainty which the indi-

vidual consciousness has of itself. Here, too, the fundamental

fact of doubt is of authoritative importance ; in this case, also, it

already contains implicitly the full truth. How should Ave come

to question and doubt the perceptions of the external world which

force themselves upon us with such elementary power, asks Augus-

tine, if we did not possess, besides these, and from other sources,

criteria and standards of truths by which to measure and examine

these perceptions ? He who doubts must know the truth, for only

for its sake does he doubt. 1 In reality, continues the philosopher,

man possesses, besides sensation (sensus), the higher capacity of

reason {intellectus, ratio), i.e. of the immediate perception of incor-

poreal truths

;

2 under the latter Augustine understands, not only

the logical laws, but also the norms of the good and the beautiful

;

in general, all those truths not to be attained by sensation, which

are requisite to elaborate and judge what is given,— the principles

of judging. 3

1 De Ver. Bel. 39, 72 f.

2 Aspectus animi, quo per se ipsum non per corpus verum intuetur : De Trin.

XII. 2, 2. Cf. Contra Acad. III. 13, 29.3 The apprehension of these intelligible truths by human consciousness was

at the first designated by Augustine quite Platonically ävä/j.vv<rt.s. It was ortho-

dox scruples against the assumption of the pre-existence of the soul that led

him to regard the reason as the intuitive faculty for the incorporeal world. Cf.

also J. Stortz, Die Philosophie des hl. Augustinus (Freiburg i. E. 1882).

Page 299: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 22.] Metaphysics of Inner Experience : Augustine. 279

Such norms of reason assert themselves as standards of judg-

ment in doubt as in all activities of consciousness ; but they

transcend, as something higher, the individual consciousness into

which they enter in the course of time : they are the same for all

who think rationally, and experience no alteration in this their

worth. Thus the individual consciousness sees itself attached in its

own function to something universally valid and far reaching. 1

But it belongs to the essence of truth that it is or exists. Augus-

tine also proceeds from this fundamental conception of the ancient,

as of every naive theory of knowledge. But the Being or existence

of those universal truths, since they are absolutely incorporeal in

their nature, can be thought only as that of the Ideas in God—after the Neo-Platonic mode ; they are the changeless Forms and

norms of all reality (principales formal vel rationes rerum stabiles

atqne incommutabiles, quo3 in divino intellectu continentur), and the

determinations of the content of the divine mind. In him they

are all contained in highest union ; he is the absolute unity, the all-

embracing truth ; he is the highest Being, the highest Good, perfect

Beauty (unum, verum, bonum). All rational knowledge is ulti-

mately knowledge of God. Complete knowledge of God, indeed,

even according to Augustine's admission, is denied to human insight

in the earthly life. Perhaps only the negative element in our idea

of him is completely certain ; and, in particular, we have no ade-

quate idea of the way in which the different elements of divine

truth which the reason beholds are united in him to form the

highest real unity. For his incorporeal and changeless essence

(essentia) far transcends all forms of relation and association that

belong to human thought ; even the category of substance applies

to him as little as do the rest.2

3. Directly consistent as these thoughts are with Neo-Platonism,?

their Christian character is yet preserved in Augustine's presenta-

tion by the fact that the religious idea of the deity as absolute

personality is inseparably fused with the philosophical conception

of the deity as the sum and essence of all truth. But just for this

reason the whole Augustinian metaphysics is built up upon the

1 De Lib. Arb. II. 7 ff.

2 The essential thing in this is the insight, that the categories acquired in

knowing Nature are inadequate for the peculiar nature of spiritual synthesis

(according to which the divine essence should be thought). The new categories

of internality are, however, with Augustine only in the process of coming into

existence ; cf . the following.3 In fact, Augustine seeks throughout to identify the vous of Plotinus with the

\6yos of Origen; but by dropping from the Neo-Platonic doctrine the emanistic

derivation of the voüs and its acquirement of independent existence, he abrogatesthe physical schema of the world potencies in favour of the psychical.

Page 300: Windelband, History of Philosophy, Vol. 1

280 Mediaeval Philosophy : First Period. [Part III.

self-knowledge of the finite personality ; that is, upon the fact of

inner experience. For so far as a comprehension of the divine

essence is at all possible for man, it can be gained only after the

analogy of human self-knowledge. This, however, shows the fol-

lowing fundamental composition of the inner life : the permanent

existence of spiritual Being is given in the sum-total of its content

of consciousness, or reproducible ideas; its movement and living

activity consists in the processes of uniting and separating these

elements in judgments ; and the impelling force in this motion is

the will, directed toward the attainment of highest blessedness.

Thus the three aspects of psychical reality are idea ( Vorstellung),

judgment, and will : memoria, intellectus, voluntas, 1 and Augustine is

expressly on his guard against conceiving of these modes of func-

tioning which are peculiar to personality, as the properties of

bodies are conceived. Just as little do they mean different strata

or spheres of its existence ; they form iu their indissoluble unity

the substance of the soul itself. In accordance with these relations

thus recognised in man's mental life, Augustine then not only seeks

to gain an analogical idea of the mystery of the Trinity, but recog-

nises, also, in the esse, nosse, and velle the fundamental determina-

tions of all reality. Being, knowing, and willing comprise all

reality, and in omnipotence, omniscience, and perfect goodness, the

deity encompasses the universe.

The outspoken opinion of the inadequacy of the physical (Aristotelian)

categories reminds us only seemingly of Neo-Platonism, whose intelligible cate-

gories (cf. p. 245), as well as its entire metaphysical schema, are throughoutphysical. It is Augustine who is first in earnest in the attempt to raise the

peculiar forms of relation characteristic of the inner nature, to metaphysicalprinciples. Aside from this, his cosmology runs on in the track laid by Neo-Platonism without peculiarities worthy of mention. The doctrine of the twoworlds, with its anthropological correlates, forms here the presupposition.

The world of sense is known through perceptions, the intelligible world through

the reason, and these two given constituents of knowledge are brought into

relation with each other by intellectual thought (ratiocinatio). For apprehend-ing Nature, the teleology conditioned by the doctrine of Ideas presents itself.

The corporeal world also is created out of nothing by divine power, wisdom, andgoodness, and bears in its beauty and perfection the sign of its origin. Evil

(including moral evil, yet cf. below) is here, too, nothing properly real ; it is

not a thing, but an act; it has no causa efficients, but only a causa deficiens

;

its origin is to be sought not in the positive Being (God), but in the lack of

Being of finite natures ; for these latter, as having been created, possess only

a weakened and therefore a defective reality. Augustine's thcjdicy stands thus

essentially upon the ground of that of Origen and Plotinus.

4. A farther and essential consequence of placing philosophy

upon a consciously anthropological basis is, in Augustine's case, thö

central position which he assigned in his theory of the universe to

1 The same triple division of the psychical activities is found among the

Stoics. Cf. rj. 187.

Page 301: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 22.] Metaphysics of Inner Experience : Augustine. 281

the will. The leading motive in this is doubtless the man's own

experience; himself a nature ardent and strong in will, as he exam-

ined and scrutinised his own personality he came upon the will as

its inmost core. On this account the will is for him the essential

element in all : omnes nihil aliud quam voluntates sunt.

In his psychology and theory of knowledge this is shown especially

in the fact that he seeks to set forth on all sides the controlling

position of the will in the entire process of ideation and knowledge. 1

While with reference to sense perception the Neo-Platonists had

distinguished between the state of corporeal stimulation and the

becoming conscious of the same, Augustine demonstrates by an

exact analysis of the act of seeing, that this becoming conscious is

essentially an act of will (intentio animi). And as physical atten-

tion is accordingly a matter of the will, so too the activity of the

inner sense (sensus interior) shows a quite analogous dependence

upon the will. Whether we bring our own states and actions as

such to our consciousness or not, depends as truly upon voluntary

reflection as does the intentional consideration of something which

belongs to our memory, and as does the activity of the combining

fantasy when directed toward a definite goal. Finally, the thinking

of the intellect (ratiocinatio) , with its judging and reasoning, is

formed completely under the direction of the purposes of the will

;

for the will must determine the direction and the end according to

which the data of outer or inner experience are to be brought under

the general truths of rational insight.

In the case of these cognitions of rational insight the relation

assumes a somewhat more involved form, for in its relation to this

higher divine truth the activity of the human mind cannot be given

the same play as in the case of its intellectual relation to the outer

world and to its own inner world. This is true even on philosophi-

cal grounds, for according to the fundamental metaphysical scheme

the active part in the causal connection must belong to the more

universal as the higher and more efficient Being (Sein). The rela-

tion of the human mind to this truth, which is metaphysically its

superior, can in the main be only a passive one. The knowledge of

the intelligible world is for Augustine also, essentially— illumination,

revelation. Here, where the mind stands in the presence of its crea-

tor, it lacks not only the creative, but even the receptive initiative.

Augustine is far from regarding the intuitive knowledge of the

intelligible truths as possibly an independent production of the

1 Cf. principally the eleventh book of the treatise De Trinitate, and besides,

especially W. Kahl, Die Lehre vom Primat des Willens bei Augustinus, DunsScotus und Descartes (Strassburg, 188ü).

Page 302: Windelband, History of Philosophy, Vol. 1

282 Mediaeval Philosophy : First Period. [Part III.

mind out; of its own nature ; indeed, he cannot even ascribe to it the

same spontaneity of attention or of directing its consciousness

(intentio) that he ascribes to the empirical cognitions of outer and

inner perception : he must, on the contrary, regard the illumination

of the individual consciousness by the divine truth as essentially an

act of grace (cf. below), in the case of which the individual con-

sciousness occupies an expectant and purely receptive attitude.

These metaphysical considerations, which might also have been

possible upon the basis of Neo-Platonism, experience in Augustine's

case a powerful reinforcement by the emphasis which he laid in his

theology upon the divine grace. Knowledge of the truths of reason

is an element in blessedness, and blessedness man owes not to his

own will, but to that of God.

Nevertheless Augustine here, too, sought to save a certain co-

operation for the will of the individual, at least at first. He not

only emphasises that God bestows the revelation of his truths upon

him only, who through good endeavour and good morals, i.e. through

the qualities of his will, shows himself a worthy subject for this

revelation ; he teaches also that the appropriation of divine truth is

effected not so much by insight, as through faith or belief. Faith

or belief, however, as ideation plus assent, though without the act

of conception, presupposes indeed the idea of its object, but contains

in the factor of assent, which is determined by no intellectual com-

pulsion, an original volitional act of the affirming judgment. The

importance of this fact extends so far, in Augustine's opinion, that

not only in divine and eternal things, but also in the human and

earthly and temporal things, this conviction produced immediately

by the will yields the original elements of thought. The insight

which conceives and comprehends grows out of these elements by

means of the combining reflective procedure of the understanding.

Thus even in the most important things, i.e. in questions of salva-

tion, faith in the divine revelation and in its appearance in the tradi-

tion of the Church— faith dictated by the good will— must precede

the knowledge which appropriates and comprehends it intellectually.

Full rational insight is indeed first in dignity, but faith in revelation

is the first in time.

5. In all these considerations of Augustine, the central point

is the conception of the freedom of the will, as a decision, choice, or

assent of the will, independent of the functions of the understand-

ing, not conditioned by motives of cognition, but rather determining

these motives without grounds in consciousness for its acts, and

Augustine faithfully exerted himself to maintain this conception

against various objections. In addition to the consciousness of

Page 303: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 22.] Metaphysics of Inner Experience : Augustine. 283

ethical and religious responsibility, it is principally the cause of

the divine justice that he here aims to defend : and, on the other

hand, most of his difficulties arise from the attempt to unite un-

caused action whose opposite is alike possible and objectively think-

able, with the divine prescience. He helps himself here by appealing

to the distinction between eternity (timelessness) and time. In an

extremely acute investigation x he maintains that time has real sig-

nificance only for the functions of inner experience as they measure

and compare : its significance for outer experience also arises only

in consequence of this. The so-called foreknowledge of the deit}',

which is in itself timeless, has as little causally determining power

for future events as memory has for those of the past. In these

connections, Aristotle is justly regarded as one of the most zealous

and forcible defenders of the freedom of the will.

But in opposition to this view, championed essentially with the

weapons of former philosophy, there now appears in Augustine's

system another line of thought, increasing in force from work to

work, which has its germ in the conception of the Church and in

the doctrine of its redeeming power. Here the principle of histor-

ical universality encounters victoriously the principle of the abso-

lute certainty of the individual mind. The idea of the Christian

Church, of which Augustine was the most powerful champion, is

rooted in the thought that the whole human race is in need of re-

demption. This latter idea, however, excludes the completely unde-

termined freedom of the will in the individual man ; for it requires

the postulate that every individual is necessarily sinful, and therefore

in need of redemption. Under the overpowering pressure of this

thought, Augustine set another theory by the side of his theory of

freedom of the will which was so widely carried out in his philo-

sophical writings ; and this second theory runs counter to the first

throughout.

Augustine desires to solve the question as to the origin of evil,

which is so important for him personally, and to solve it— in

opposition to Manichaeism— by the conception of the freedom of

the will, in order to maintain in this, human responsibility and

divine justice ; but in his theological system it seems to him to be

sufficient to restrict this freedom of will to Adam, the first man.

The idea of the substantial oneness of the human race— an idea

which was a co-operating element in the faith in the redemption of

all by the one Saviour— permitted likewise the doctrine that in

1 In the eleventh book of the Confessions. Cf. C. Fortlage, A. De TemporeDoctrina (Heidelberg, 1836).

Page 304: Windelband, History of Philosophy, Vol. 1

284 Mediceval Philosophy : First Period. [Part III.

the one man Adam all humanity had sinned. By the abuse of this

freedom of the will on the part of the first man, the whole humannature has been so corrupted that it cannot do otherwise than sin

{non posse non peccare) . This loss of freedom applies without ex-

ception, to the whole race arising from Adam. Every man brings

with him into the world this corrupted nature which is no longer

capable of good in its own strength or freedom, and this inherited

sin is the punishment for original sin. Just from this it follows

that all men, without exception, are in need of redemption and of

the Church's means of grace. One as little as another deserves to

receive this grace : therefore, thinks Augustine, no injustice can

be seen in the fact that God bestows this grace, to which no one

has any claim, not upon all, but only upon some ; and it is never

known upon whom. But, on the other hand, the divine justice

demands that, at least in the case of some men, the punishment for

Adam's fall should be permanently maintained, that these men,

therefore, should remain excluded from the working of grace and

from redemption. Since, finally, in consequence of their corrupted

nature, all are alike sinful and incapable of any improvement of

themselves, it follows that the choice of the favoured ones takes

place not according to their worthiness (for there are none worthy

before the working of grace), but according to an unsearchable

decree of God. Upon him whom he will redeem he bestows his

revelation with its irresistible power : he whom he does not choose,

— he can in nowise be redeemed. Man in his own strength cannot

make even a beginning toward the good : all good comes from Godand only from him.

In the doctrine ofpredestination, accordingly (and this is its philo-

sophical element), the absolute causality of God suppresses the free

will of the individual. The latter is refused both metaphysical

independence and also all spontaneity of action ; the individual is

determined either by his nature to sin or by grace to the good. So

in Augustine's system two powerful streams of thought come into

violent opposition. It will always remain an astonishing fact that

the same man who founded his philosophy upon the absolute and

independent certainty of the individual conscious mind, who threw

the plummet of the most acute examination into the depths of inner

experience and discovered in the will the vital ground of spiritual

personality, found himself forced by the interests of a theological

controversy to a theory of the doctrine of salvation which regards

the acts of the individual will as unalterably determined conse-

quences, either of a general corruption or of the divine grace.

Individualism and universalism in the conception of psychical reality

Page 305: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 22.J Metaphysics of Inner Experience : Augustine, 285

stand here in bald opposition, and their clashing contradiction is

scarcely concealed by the ambiguity of the word "freedom," which,

in the one line, is defended according to its psychological meaning,

in the other, according to its ethico-religious meaning. The oppo-

sition, however, of the two motives of thought which here lie side by

side so irreconcilable, had influence in the succeeding development

of philosophy until long past the Middle Ages.

6. In the light of the doctrine of predestination the grand picture

of the historical development of humanity, which Augustine drew

in the manner and spirit of the old patristic philosophy, takes on

dark colours and peculiarly stiff, inflexible forms. For if not only

the course of the history of salvation taken as a whole, but also,

as in Augustine's system, the position which every individual is to

occupy within it, has been previously fixed by divine decree, one

cannot rid one's self of the gloomy impression that all man's voli-

tional life in history, with all its thirst for salvation, sinks to a

play of shadows and puppets, whose result is infallibly fixed from

the beginning.

The spiritual world throughout the whole course of history falls

apart, for Augustine, into two spheres,— the realm of God and the

realm of the devil. To the former belong the angels that have not

fallen, and the men whom God has chosen for his grace ; the other

embraces, together with the evil demons, all those men who are not

predestined to redemption, but are left by God in the state of sin and

guilt : the one is the kingdom of heaven, the other that of the world.

The two occupy in the course of history a relation like that of two

different races which are mingled only in outer action, while in-

ternally they are strictly separate. The community of the elect has

no home on earth ; it lives in the higher unity of divine grace. Thecommunity of the condemned, however, is divided within itself bydiscord ; it fights in earthly kingdoms for the illusory worth of

power and rule. Christian thought at this stage of development

is so little able to master the reality presented by the world, that

Augustine sees in the historical states only the provinces of a com-

munity of sinners in hostility to God, condemned to quarrel with

one another. For him, in fact, the kingdom of God is still not of

this world ; and the Church is for him the saving institution of the

divine kingdom, which enters the temporal life.

The course of the world's history under these presuppositions

is so conceived that we find a division entering between the tworealms, which becomes sharper and sharper in the course of history,

and ultimately results in the complete and definitive separation of

the same. In six periods, which correspond to the creative davs of

Page 306: Windelband, History of Philosophy, Vol. 1

286 Medieval Philosophy : First Period. [Part III

the Mosaic cosmogony and are attached to dates of Israelitic his-

tory, Augustine constructs his history of the world. In this process

he combines a depreciatory estimate of the Roman world with slight

understanding of the essential nature of the Grecian. The decisive

point in this development is for him, also, the appearance of the

Saviour, by which not only the redemption of those chosen by grace

is brought to completion, but also their separation from the children

of the world. With this begins the last Avorld-period, whose end will

be the Judgment : then after the stress of conflict shall enter the Sab-

bath, the peace of the Lord— but peace only for the elect ; for those

not predestined to salvation will then be completely separated from

the saints, and entirely given over to the pain of their unhappiness.

However spiritually sublime (though never without attendant

physical imagery) the conception of happiness and pain here pre-

sented,— and this sublimity is especially noteworthy in the thought

of unhappiness as a weakening of Being, due to the lack of divine

causality,— the dualism of the Good and the Evil is yet unmistak-

ably, for Augustine, the final issue of the world's history. The manassailed by so many powerful motives of thought has not overcome

the Manichveism of his youthful belief; he has taken it up into

Christian doctrine. Among the Manichaeans the antithesis of

good and evil is held to be original and indelible : with Augustine

this antithesis is regarded as one that has come into being, and yet

as one that is ineradicable. The omnipotent, omniscient, supremely

benevolent God has created a world which is divided forever into

his own realm and that of Satan.

7. Among the complicated problems and ideas of universal his-

torical importance which Augustinianism contains, there is still one

to be brought forward. It lies in the conception of blessedness itself

in which all motives of his thought cross. For, strongly as Augus-

tine recognised in the will the inmost motive energy of humannature, deeply as he penetrated the striving after happiness as the

impelling motive of all psychical functions, he yet remained firmly

convinced that the satisfaction of all this stress and urging is to be

found only in beholding divine truth. The highest good is God ; but

God is the truth, and one enjoys truth by beholding it and resting in

its contemplation. All urging of the will is but the path to this

peace in which it ceases. The last task of the will is to be silent in

the gracious working of divine revelation,— to remain quiet when

the vision of truth, produced from above, comes over it.

Here are united in common opposition to individualism of will,

the Christian idea of the absolute causality of God, and the contem-

plative mysticism of the Neo-Platonists. From both sides, the same

Page 307: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 23.] Controversy over Universalis. 287

tendency is at work to bring about the conception of man's sanctifi-

cation as a working of God in him, as a becoming filled and illumined

by the highest truth, as a will-less contemplation of the one, infinite

Being. Augustine, indeed, worked out forcibly the practical conse-

quences which the working of grace should have in the earthly life,

purification of the disposition and strictness in the conduct of life,—and just in this is shown the comprehensive breadth of his personal

nature and his spiritual vision. He develops the vigorous energy of

his own combative nature into an ethical doctrine, which, far re-

moved from the asceticism of Neo-Platonism with its weariness of

life, sets man in the midst of the world-battle between Good and

Evil as a brave fighter for the heavenly kingdom. But the highest

reward which beckons this fighter for God is yet, for Augustine, not

the restless activity of the will, but the rest of contemplation. For

the temporal life, Augustine demands the full and never-resting

exertion of the struggling and acting soul; for eternity he offers the

prospect of the peace of becoming absorbed in divine truth. Heindeed designates the state of the blessed as the highest of the

virtues, as love 1 (charitas), but in the eternal blessedness where the

resistance of the world and of the sinful will is no longer to be over-

come, where love has no longer any want that must be satisfied,

there this love is no longer anything other than a God-intoxicated

contemplation.

In this duality, also, of the Augustinian ethics, old and new lie

close together. With the tense energy of will which is demanded

for the earthly life, and with the transfer of the ethical judgment

so as to make it apply to the inner disposition, the modern manappears ; but in the conception of the highest goal of life the ancient

ideal of intellectual contemplation retains the victory.

Here lies in Augustine's doctrine itself a contradiction with the

individualism of the will, here at a decisive point an Aristotelian,

Neo-Platonic element maintains itself, and this internal opposition

unfolds itself in the formation of the problems of the Middle Ages.

§ 23. The Controversy over TJniversals.

Johannes Saresberiensis, 3Ietalogicus, II. cap. 17 f.

J. H. Löwe, Der Kampf zwischen Nominalismus und Realismus im Mittel-

älter, sein Ursprung und sein Verlauf (Prague, 1876).

The schooling in formal logic which the peoples that entered

upon the scientific movement at the beginning of the Middle Ages

1 In his system the three Christian virtues, faith, hope, and love, are placed

above the practical and dianoetic virtues of Greek ethics.

Page 308: Windelband, History of Philosophy, Vol. 1

288 Mediceval Philosophy : First Period. [Tart III

were obliged, to undergo, developed in connection with the question

as to the logical and metaphysical significance of genera and species

(universalia) . But it would be a grave mistake to suppose that this

question had only the didactic value of serving as a subject for

mental drill, in connection with which the rules of conceptional

thought, division, judgment, and inference, were impressed for cen-

turies upon ever new and increasing throngs of scholars. On the

contrary, the tenacity with which the science of the Middle Ages—and it is significant that this occurred independently in the Orient

as well as in the Occident— held fast to the elaboration of this

problem in endless discussions, is rather in itself a proof that in this

question a very real and very difficult problem lies before us.

In fact, when Scholasticism, in its timorous beginnings, made the

passage in Porphyry's Introduction 1 to the Categories of Aristotle

which formulated this problem, the starting-point of its own first

attempts at thought, it hit with instinctive sagacity upon precisely

the same problem which had formed the centre of interest during

the great period of Greek philosophy. After Socrates had assigned

to science the task of thinking the world in conceptions, the ques-

tion how the class-concepts, or generic conceptions, are related to

reality, became, for the first time, a chief motive of philosophy. It

produced the Platonic doctrine of Ideas and the Aristotelian logic

;

and if the latter had as its essential content (cf. § 12) the doctrine

of the forms in which the particular is dependent upon the uni-

versal, it is easy to understand that even from so scanty remains

and fragments of this doctrine as were at the service of the earliest

Middle Ages, the same problem must arise with all its power for the

new race also. And it is likewise easy to understand that the old

enigmatic question worked upon the naive minds of the Middle

Ages, untrained in thought, in a manner similar to that in which it

worked upon the Greeks. In fact, the delight in logical dispute, as

this developed after the eleventh century at the schools of Paris,

finds its counterpart as a social phenomenon only in the debates of

the philosophers at Athens, and in these latter, too, as numerous

anecdotes prove, the question as to the reality of universals, which

was connected with the doctrine of Ideas, played a leading part.

Nevertheless the problem was renewed under conditions that

were essentially less favourable. When this question emerged for

the Greeks, they possessed a wealth of proper scientific experience

1 The formulation of the problem in the translation of Boethius is as follows

:

"... de generibus et speciebus— sive subsistant sive in solis nudis intellectibus

posita sint, sive subsistentia corporalia an incorporalia, et utrum separata asensibilibus an in sensibilibus posita et circa /jcec consistentia. . .

."

Page 309: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 23.] Controversy over Universale : Scotus Erigena. 289

and a store of real information and knowledge, which, if not always,

yet for the most part and on the whole, prevented them from mak-

ing their discussion solely a game with the abstractions of formal

logic. But mediaeval science, especially in its beginnings, lacked

just this counterpoise, and on this account was obliged to move so

long in a circle with the attempt to construct its metaphysics out

of purely logical considerations.

That the Middle Ages, in their turn, engaged and persisted so

pertinaciously in this controversy which had previously been waged

principally between Plato and the Cynics, and afterward between

the Academy, the Lyceum, and the Stoa, was not due solely to the

fact that in consequence of the defective character of their tradi-

tions the thinkers of the Middle Ages knew as good as nothing of

those earlier debates ; it had yet a deeper ground. The feeling of

the peculiar, intrinsic worth of personality, which had gained so

powerful expression in Christianity and especially in the Augustin-

ian doctrine, found the liveliest echo and the strongest sympathy

among precisely those tribes which were called to become the newbearers of civilisation; and in the hearts of these same peoples

surged also the youthful delight in richly coloured reality, in the

living, particular appearance. But with the Church doctrine they

received a philosophy which, with the measured calm of Greek

thought, conceived the essential nature of things to lie in universal

connections, a metaphysics which identified the stages of logical

universality with intensities of Being of varying worths. In this

lay an inconsistency which covertly asserted itself, even in Augus-

tinianism, and became a constant stimulus for philosophical reflec-

tion.

1. The question as to the individual's ground of Being or exis-

tence, from which mediaeval thought never became free, was the

more natural for it just at its beginning in proportion as the Neo-

Platonic metaphysics still maintained itself under the veil of a

Christian mysticism. Nothing could be more adapted to call out

the contradiction of a natural individualism than the high degree

of consistency with which Scotus Erigena carried through the funda-

mental thoughts of the Neo-Platonic Realism. Perhaps no philoso

pher has expressed more clearly and frankly than he the fina?

consequences of the metaphysics which, from the standpoint of the

Socratic-Platonic principle that the truth, and therefore also Being,

is to be sought in the universal, identifies the stages of universality

with those of the intensity and priority of Being. The universal

(the class-concept or logical genus) appears here as the essential and

original reality, which producesfrom itself and contains within itselj

Page 310: Windelband, History of Philosophy, Vol. 1

290 Mediaeval Philosophy : First Period. [Part IIL

the particular (the species and ultimately the individual). The

universals are, therefore, not only substances (?*es; hence the name

" Realism"), but, as contrasted with the corporeal individual things,

they are the more primitive, the producing and determining sub-

stances ; they are the more Real substances, and they are the more

Real in proportion as they are the more universal. In this conception,

therefore, the logical relations of concepts immediately become

metaphysical relations ; formal arrangement contains real signifi-

cance. Logical subordination becomes changed into a production

and inclusion of the particular by the general ; logical partition and

determination become transformed into a causal process by means

of which the universal takes on form and unfolds itself in the

particular.

The pyramid of concepts, thus raised to a metaphysical signifi-

cance, culminates in the concept of the deity as the most universal.

But the last product of abstraction, the absolutely universal, is that

which has no determinations (cf. p. 250). Hence this doctrine

becomes identical with the old "negative theology," according to

which we can predicate of God only what he is not

;

l and yet here,

too, this highest Being is designated, quite in accord with the

thought of Plotinus, as the "uncreated, but self-creating Nature."

For this most universal Being produces out of itself all things;

these, therefore, contain nothing else than its manifestations, and

are related to it as particular specimens or instances are to the

class ; they are in it and exist only as its modes of appearance.

The result of these presuppositions is thus a logical pantheism : all

things of the world are " theophanies "; the world is God developed

into the particular, proceeding out of himself to take on a definite

form (deus explicitus). God and the world are one. The same

"Nature" (<£v<ns) is, as creative unity, God, and as created plurality,

the world.

The process of unfolding (eg7*essus) proceeds in the graded scale

of logical universality. Out of God comes at first the intelligible

world as "the Nature which is created and itself creates," the realm

of universals, of Ideas which (as vol in the sense of Plotinus) form

the working forces in the sensuous world of phenomena. The

Ideas are built up as a heavenly hierarchy according to their various

grades of universality, and therefore also of intensity of Being, and

in connection with this thought Christian Mysticism constructs a

1 In carrying out this Philonic thought (cf. p. 237) the Church Fathers hadalready employed a course of thought which proceeds by successive abstraction

to the concept of God as the undetermined. Cf., e.g., Clement Alex. Strom.V. 11 (683).

Page 311: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 23.] Controversy over Universalis : Scotus Eriyena. 291

doctrine of angels after a Neo-Platonic pattern. But in every case

beneath the mythical covering the important thought is really

active, that real dependence consists in logical dependence; the

logical consequence, by which the particular follows from the

general, is spuriously substituted for the causal relation.

Hence, then, even in the world of the senses, it is only the uni-

versal that is properly active and efficient : corporeal things, as a

whole, form the " Nature which is created and does not itself

create." * In this world the individual thing is not as such active

;

it is rather active according to the proportion of universal attri-

butes which attain manifestation in it. The individual thing of

sense, accordingly, possesses the least force of Being, the weakest

and completely dependent species of reality : the Neo-Platonic

Idealism is maintained by Scotus Erigena in full.

To the stages of unfolding corresponds in a reverse order the

return of all things into God (regressus), the resolution of the

world of individual forms into the eternal primitive Being, the dei-

fication of the world. So thought, as the final goal of all genera-

tion and change, as the extinction of all that is particular, God is

designated as " the Nature which neither is created nor creates "

:

it is the ideal of motionless unity, of absolute rest at the end of the

world-process. All theophanies are destined to return into the

unity of the divine All-Being,— that unity which knows no dis-

tinctions. Thus, even in the final destiny of things, the superior

reality of the universal, which swallows up all that is particular,

preserves itself.

2. As in antiquity (cf. § 11, 5), so here, in consequence of the

effort to assure truth and reality to universals, the peculiar thought

of a graded scale of Being appears. Some things (universals), is

the doctrine, are more than others (particulars). " Being" is looked

upon as, like other qualities, capable of comparison, of increase and

diminution ; it belongs to some things more than to others. So it

became the custom to think that the concept of Being (esse, existere)

has a relation to that which is (essentia), and a relation of different

degrees of intensity, just as other marks and qualities are related

to the objects in which they are formed. As a thing possesses more

or less extension, force, permanence, so it has also more or less

" Being "; and as it can receive or lose other qualities, so it can

receive or lose that of Being. This line of thought, peculiar to

Eealism, must be kept in mind to understand a great number of the

1 It need only be briefly mentioned that this "division of Nature" obviouslyrecalls the Aristotelian distinction of the unmoved mover, the moved mover,and that which neither moves nor is moved. Cf. § 13, 5.

Page 312: Windelband, History of Philosophy, Vol. 1

292 Mediceval Philosophy : First Period. [Part III.

metaphysical theories of the Middle Ages. It explains, in the first

place, the most important doctrine which Eealism produced, the

ontological argument for the existence of God which Anselm of Can-

terbury brought forward.

The more universality, the more Reality. From this it follows

that if God is the most universal being, he is also the most Real

;

if he is the absolutely universal being, he is also the absolutely Real

being, ens realissimum. He has, therefore, according to the concep-

tion of him, not only the comparatively greatest Reality, but also

absolute Reality ; that is, a Reality than which a greater and higher

cannot be thought.

But through the whole development which this line of thought

had already taken in antiquity, we find that the worth-predicate of

perfection was inseparably fused with the conception of Being.

The degrees of Being are those of perfection ; the more anything

is, the more perfect it is, and, vice versa, the more perfect anything

is, the more it is.1 The conception of the highest Being is, there-

fore, also that of an absolute perfection ; that is, of a perfection such

that it cannot be thought higher and greater : ens perfectissimwni

In accordance with these presuppositions, Anselm is perfectly

correct in his conclusion that, from the mere conception of God as

most perfect and most real Being, it must be possible to infer his

existence. But to do this he attempts various modes of proof. In

his Monologium he follows the old cosmological argument that

because there is Being at all, a highest and absolute Being must

be assumed from which all else that exists has its Being, and which

itself exists only from itself, according to its own essential nature

(aseitas). Whereas every individual existent entity can be also

thought as non-existent, and therefore owes the reality of its essence

not to itself, but to another (the Absolute), the most perfect Being

can be thought only as being or existent, and exists accordingly

only by virtue of the necessity of its own nature. God's essence

(and only God's) involves his existence. The nerve of this argu-

ment is thus ultimately the Eleatic basal thought, lanv ehai, Being

is, and cannot be thought otherwise than as being or existing.

Anselm, however, involved this same thought in a peculiar com-

plication, while he intended to simplify it and render it independent

in itself. In the Proslogium he entered upon the ontological argu-

ment, properly so called, which maintains that without any reference

to the Being of other things, the mere conception of the most per-

1 A principle which lies at the basis of Augustine's theodicy, m so far as withboth the existent is held to be eo ipso good, and the evil on the contrary, as nottruly existent.

Page 313: Windelband, History of Philosophy, Vol. 1

CaAP. 1, § 23.] Controversy over Universals : Anselm. 293

feet Being involves its Reality. Inasmuch as this conception is

thought, it possesses psychical reality : the most perfect being is as

a content in consciousness {esse in intellectu). But if it existed

only as a content in consciousness, and not also in metaphysical

reality (esse etiam in re), a still more perfect being could evidently

be thought, which should possess not only psychical, but also meta-

physical reality ; and thus the former would not be the most perfect

being possible. It belongs, accordingly, to the conception of the

most perfect being (quo majus cogitari non jiotest) that it possesses

not only reality in thought, but also absolute reality.

It is obvious that Anselm in this formulation was not fortunate

in his shift, and that what hovered before him attained in this

proof but a very awkward expression. For it takes little acuteness

to see that Anselm proved only that if God is thought (as most

perfect being), he must be thought also necessarily as being or

existent, and cannot be thought as non-existent. But the ontologi-

cal argument of the Proslogium did not show even in the remotest

degree that God, i.e. that a most perfect being, must be thought.

The necessity for this stood fast for Anselm personally, not only

because of the conviction of his faith, but also by the cosmological

argumentation of the Monologium. When he believed that he

could dispense with this presupposition and with the help of the

mere conception of God arrive at the proof of his existence, he

exemplified in typical manner the fundamental idea of Realism,

which ascribed to conceptions without any regard to their genesis

and basis in the human mind, the character of truth, i.e. of

Reality. It was on this ground alone that he could attempt to

reason from the psychical to the metaphysical reality of the concep-

tion of God.

The polemic of Gaunilo, therefore, in a certain respect hit the

vulnerable point. He argued that according to the methods of

Anselm, in quite the same manner the reality of any idea whatever,

e.g. that of an island, if the mark of perfection were only included

within it, might be proved. For the most perfect island, if it were

not really in existence, would evidently be surpassed in perfection

by the real island, which should possess the same other marks ; the

former would be inferior to the latter in the attribute of Being.

But instead of showing in his rejoinder, as might have been ex-

pected, that the conception of a perfect island is a completely unnec-

essary arbitrary fiction, or that this conception contains an inner

contradiction, while the conception of the most real being is neces-

sary and not contradictory, Anselm expatiates further upon his

argument, that if the most perfect being is in the intellect, it must

b°- also in r€.

Page 314: Windelband, History of Philosophy, Vol. 1

294 Mediaeval Philosophy : First Period. [Part III.

However slight the cogency of this attempted proof remains for

him who does not, as Anselm does without acknowledging it, regard

the conception of an absolute Being as a necessity of thought, the

ontological argument is yet valuable as the characteristic feature of

mediaeval Realism, of which it forms the most consistent expression.

For the thought that the highest being owes its reality only to its

own essential nature, and that therefore this reality must be capable

of being proved from its conception alone, is the natural conclusion

of a doctrine which traces the Being of things of perception back to

a participation in conceptions, and again within the conceptions

themselves sets up a graded scale of reality, employing the degree

of universality as the standard.

3. When now the question arose as to the kind of reality which

belongs to universals, and as to their relation to the individual

things known to the senses, mediaeval Realism found itself involved

in difficulties quite similar to those which had faced the Platonic

Realism. The thought of a second, higher, immaterial world, which

at that former period had to be born, was now indeed received as a

complete and almost self-evident doctrine, and the religiously dis-

posed thinking could be only sympathetic in its attitude toward the

JSTeo-Platonic conception of the Ideas as contents of the divine mind.

Following the pattern of the Platonic Timaeus, whose mythical mode

of presentation was favourable to this conception, Bernard of

Chartres sketched an imaginative cosmogonic work of fantastic

grotesqueness, and we find with his brother Theodoric, attempts, sug-

gested by the same source, to construct a symbolism of numbers,

which undertook not only, as was done in other instances, to develop

the dogma of the Trinity, but also to develop further fundamental

metaphysical conceptions out of the elements of unity, likeness, and

unlikeness. 1

In addition to this question concerning the archetypal reality of

the Ideas in the mind of God, the question is also, what significance

is to be conceded to them in the created world. Extreme Realism,

as it had been maintained at the outset by William of Ghampeaux,

taught the full substantiality of the class-concept in this world also

;

the universal is present in all its individuals as the undivided

essence, everywhere identical with itself. The class accordingly

appears as the unitary substance, and the specific marks of thb indi-

viduals belonging to it appear as the accidents of this substance.

It was Abelard's objection that according to this theory mutually

contradictory accidents would have to be ascribed to the same sub-

: Cf. the extracts in HaurSau, Hist. d. 1. ph. sc, I. 396 ff.

Page 315: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 23.] Controversy over Universals : Realism. '295

stance, which first forced the defender of Realism to give up this

extreme position and restrict himself to the defence of the proposi-

tion, that the class exists in the individuals, individuaiiter

;

l i.e.

that its universal, identical essence clothes itself in each particular

example in a particular substantial Form. This view was in touch

with the conception of the Neo-Platonists, which had been main-

tained by Boethius and Augustine and also occasionally mentioned

in the literature of the intervening period, and its exposition movesreadily in the Aristotelian terminology, according to which the

universal appears as the more indeterminate possibility which

realises itself in individuals by means of their peculiar Forms.

The conception is then no longer substance in the proper sense, but

the common substratum which takes on different forms in individ-

ual instances.

Walter of Mortagne sought to remove the difficulty in another

way, by designating the individualising of the classes or genera to

species, and of the species to individual things, as the entering of

the substratum into different states (status), and yet regarding

these states as realiter specialising determinations of the universal.

In both these lines of thought, however, Realism was only with

difficulty held back from a final consequence which at the first lay

in nowise within the purpose of its orthodox supporters. The re-

lation of the universal to the particular might be regarded as the

self-realising of the substratum into individual Forms, or as its

specialisation into individual states,— in either case one came ulti-

mately in the ascending line of abstract conceptions to the idea of

the ens generalissimum, whose self-realisations, or whose modified

states, formed in descending line the genera, species, and individuals,

i.e. to the doctrine that in all phenomena of the world only the one

divine substance is to be seen. Pantheism inhered in the blood of

Realism by reason of its Neo-Platonic descent and was always

making its appearance here and there ; and opponents like Abelard

did not fail to cast this consequence in the face of Realism.

Meanwhile realistic pantheism did not come to be expressly

maintained in this period; on the other hand, Realism in its theory

of universals found an instrument for establishing some of the

fundamental dogmas, and therefore rejoiced in the approbation of

the Church. The assumption of a substantial reality of the logi-

cal genera not only seemed to make possible a rational exposi-

tion of the doctrine of the Trinity, but also, as was shown byAnselm and Odo (Odardus) of Cambrey, proved to be a fit phil-

1 For the reading "indifferenter," of. Löwe, op. cit., 49 ff., and CI. Bauuiker,Arch. f. Gesch. d. IJ

h., X. 257.

Page 316: Windelband, History of Philosophy, Vol. 1

296 Mediceval Philosophy : First Period. [Part III.

osophical basis for the doctrines of inherited sin and vicarious

satisfaction.

4. On the same grounds, we find at first the reverse lot befalling

Nominalism, which during this period remained more repressed and

stifled. Its beginnings 1 were harmless enough. It grew out of the

fragments of Aristotelian logic, in particular out of the treatise DeCategoriis. In this the individual things of experience were desig-

nated as the true "first" substances, and here the logico-grammatical

rule was propounded that " substance " could not be predicate in

a judgment : res non predicatur. Since now the logical significance

of universals is essentially that of affording the predicates in the

judgment, (and in the syllogism), it seemed to follow— this the

commentary Super Porphyrium had already taught— that univer-

sals could not be substances.

What are they, then ? It could be read in Marcianus Capella that

a universal was the comprehension of many particularities by one

name (normen), by the same word (vox); but a word, Boethius had

defined as a "motion of the air produced by the tongue." Withthis all elements of the thesis of extreme Nominalism were given

:

universals are nothing but collective names, common designations

for different things, sounds (flatus vocis), which serve as signs for a

multiplicity of substances or their accidents.

In what degree the thus formulated Nominalism, which in this

extreme form must have ignored even the real occasions for such

collective names, was actually propounded and defended during that

period 2 can no longer be determined.3 But the metapihysics of indi-

vidualism which corresponds to such a theory of knowledge meets us

clearly and firmly with the claim that only individual things are to

be regarded as substances, as truly real. This was doubtless most

sharply expressed by Roscellinus, when he presented it in a two-

fold aspect : as the comprehension of many individuals under the

same name is only a human designation, so, too, the distinguishing

of parts in individual substances is only an analysis for humanthought and communication

;

4 the truly real is the individual thing,

and that alone.

1 Cf. C. S. Barach, Zur Geschichte des Nominalismus vor Boscellin (Vienna,

1866).2 It is certain that this did not as yet occur in the beginnings of Nominalism

(with Eric of Auxerre, with the author of the commentary Super Porphyrium,etc.), for with these writers we find at the same time the expression of Boethiusthat genus is sxibstantialis similitudo ex diversis speciebus in cogitatione collecta.

3 John of Salisbury says (Polier. VII. 12 ; cf. Metal. II. 17) that this opinion

vanished again with its author Roscellinus.4 The example of the house and its wall, which, according to Abelard ( Ouvr.

Ined. 471), he employed in this connection, was certainly the most unfortunate

that could be thought of. How inferior such considerations are to the begin-

nings of Greek thought!

Page 317: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 23.] Controversy over Universals : Nominalism. 297

The individual, however, is that which is given in the world of

sensible reality ; hence for this metaphysics, knowledge consists

only in the experience of the senses. That this sensualism appeared

in the train of Nominalism, that there were men who allowed their

thinking to go on entirely in corporeal images, we are assured, not

only by Anselm, but also by Abelard : but who these men were

and how they carried out their theory we do not learn.

This doctrine became momentous through its application to theo-

logical questions by Berengar of Tours and Roscellinus. The one

contested, in the doctrine of the Sacrament, the possibility of the

transmutation of the substance while the former accidents were

retained ; the second reached the consequence that the three persons

of the divine Trinity were to be looked upon as three different

substances, agreeing only in certain qualities and workings (tri-

theism).

5. In the literary development of these antitheses Realism passed

current as Platonic, Nominalism as Aristotelian. The latter desig-

nation was evidently much more distorted than the former, but

when we consider the defective nature of the transmitted material,

we can understand that the mediating tendencies which thrust

themselves in between Realism and Nominalism introduced them-

selves with the endeavour to harmonise the two great thinkers of

antiquity. Of such attempts, two are chiefly worthy of mention

:

from the party of Realism the so-called Indifferentism, from that of

Nominalism the doctrine of Abelard.

As soon as Realism abandoned the doctrine of the separate

existence of the concepts (the Platonic xwPLcrlx^) an(^ supported

only the " universalia in re,-' the tendency asserted itself to con'

ceive of the different stages of universality as the real states of one

and the same substratum. One and the same absolute reality is, in

its different " status," animate being, man, Greek, Socrates. As the

substratum of these states the moderate Realists regarded the uni-

versal, and ultimately the ens realissimum; it was therefore a

significant concession to Nominalism when others made the indi-

vidual the supporter of these states. The truly existent, these

latter thinkers conceded, is the individual tiling, but the individual

thing supports within itself as essential determinations of its own

nature certain qualities and groups of qualities which it has in

common with others. This real similarity (consimilitvdo) is the

indifferent ("not different") element in all these individuals, and

thus the genus is present in its species, the species in its indi-

vidual examples, indifferenter. Adelard of Bath appears as the

chief supporter of this line of thought, vet it must have had a

Page 318: Windelband, History of Philosophy, Vol. 1

298 Mediceval Philosophy : First Period. [Part III.

wider extension, perhaps with a somewhat stronger nominalistic

accent.1

6. But it was Abelard 2 with his all-sided activity who formed

the vigorous centre in the controversy over universals. The pupil

and at the same time the opponent both of Roscellinus and of

William of Champeaux, he fought Nominalism and Realism each

by means of the other, and since he takes the weapons of his

polemic now from the one side now from the other, it could not fail

to result that his position should be interpreted and judged oppo-

sitely.3 And yet the outlines of this position are clear and dis-

tinct before us. In his polemic against all kinds of Realism, the

thought that the logical consequence of Realism is pantheism

returns so frequently and energetically that we must see in it, not

merely a convenient weapon for use in the ecclesiastical conditions

then prevailing, but rather the expression of an individualistic con-

viction easy to understand in the case of a personality so energetic,

self-conscious, and proudly self-reliant. But this individuality had

at the same time its inmost essence in clear, sharp, intellectual

activity, in genuine French rationality. Hence its no less powerful

opposition against the sensualistic tendencies of Nominalism.

Universals, Abelard teaches, cannot be things, but just as little

ean they be mere words. The word (vox) as a complex of sounds,

is indeed something singular ; it can acquire universal meaning only

mediately, by becoming a predicate (sermo) . Such an employment

of a word for a predicate is possible only through conceptional

thought (conceptus), which, by comparing the contents of percep-

tion, gains that which is by its nature adapted to become a predicate

(quod de pluribus natum est prcedicari) .

4 The universal is then the

conceptual predicate (Sermonism), or the concept itself (GoncepUml-

ism). 5 But if the universal as such gains its existence first in

thought and judgment, and in the predicate which is possible only

by this means, and exists only there, it is not therefore entirely

without relations to absolute reality. Universals could not be the

indispensable forms of all knowledge, as they in fact actually are,

if there were not something in the nature of things which we

1 According to the statements in the treatise De Generibus et Speciebus andthe communications of Abelard in his gloss on Isagoge. It seems, too, that Wil-liam of Champeaux inclined toward Indifferentism at the last.

2 Cf. S. M. Deutsch, Peter Abaelard, ein kritischer Theolog. des zwölftenJahrhunderts (Leips. 1883).

3 Thus Ritter makes him a Realist ; Haureau, a Nominalist.* Cf. Arist. De Interpr. 7, 17 a 39.5 It seems that Abelard at different times emphasised sometimes the one

alternative, sometimes the other, and perhaps his school also developed differ-

ently in accordance with these two lines of thought.

Page 319: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 23.] Controversy over Universals .• Abelard. 299

apprehend and predicate in these universals. This something is the

likeness or similarity (conformitas) of the essential characteristics

of individual substances. 1 Not as numerical or substantial identity,

but as a multiplicity with like qualities, does the universal

exist in Nature, and it becomes a unitary concept which makes

predication possible, only when it has been apprehended and con-

ceived by human thought. Even Abelard, however, explains this

likeness of character in a multiplicity of individuals upon the

hypothesis that God created the world according to archetypes which

he carried in his mind (noys). Thus, according to his view, the

universals exist firstly, before the things, as concejrtus mentis in God;

secondly, in the things, as likeness of the essential characteristics of

individuals ; thirdly, after things, in the human understanding as its

concepts and predicates acquired by comparative thought.

Thus, in Abelard the different lines of thought of the time

become united. But he had developed the individual elements of

this theory incidentally, partly in connection with his polemic, and

perhaps, also, at different times with varying emphasis on this or

that element : a systematic solution of the whole problem he never

gave. As regards the real question at issue he had advanced so far

that it was essentially his theory that became the ruling doctrine in

the formula accepted by the Arabian philosophers (Avicenna), "uni-

versalia ante multipUeitatem, in ynultijrticitate et post multiplicitatem ;"

to universals belongs equally a significance ante rem as regards the

divine mind, in re as regards Nature, and post rem as regards humanknowledge. And since Thomas and Duns Scotus in the main agreed

in this view, the problem of universals, which, to be sure, has not

yet been solved, 2 came to a preliminary rest, to come again into the

foreground when Nominalism was revived (cf. § 27).

1 Others, who in the main had the same thought, e.g. Gilbert de la Porree,

aided themselves with the Aristotelian distinction between first and secondsubstances, or between substance and subsistence

;yet Gilbert uses the latter

terms in a changed meaning as compared with their use by Abelard.2 Even if the problem as to the universals be restricted, according to the

mode of Scholasticism, to the reality of the class-concepts, the problem hasgone through essentially new phases in its further development, and cannot beregarded as finally solved by the position taken by science to-day. Behind this,

however, rises the more general and more difficult question, what metaphysicalsignificance belongs to those universal determinations, in a knowledge of whichall explanatory science practically consists. Cf. H. Lotze, Lor/ik (Leips. 1874),

§§ 313-821. [Eng. tr. ed. by B. Bosanquet, Oxford and N.Y. i«88.]

To the investigators of to-day, therefore, who would throw the controversyover universals to the lumber pile of past theories, or treat it as a long-outgrownchildren's disease, so long as they do not know how to state with completecertainty and clearness in what consists the metaphysical reality and efficiency

of that which we call a law of Nature, we must still cry, " mutato nomine de te

fabula narrata." Cf., also, O. Leibmann, Zur Analysis der Wirklichkeit (2d

ed., Strassburg, 1880), 313 ff., 471 ff., and Gedanken und Thatsachen (1 Heft,

Strassburg, 1882), 89 ff.

Page 320: Windelband, History of Philosophy, Vol. 1

300 Mediceval Philosophy : First Period. [Part IIL

7. But Abelard has a still greater significance than that due to

this central position in the controversy over universals, for he mani-

fested in his own person, and expressed in typical form, the attitude

which the dialectic, unfolding in connection with that controversy,

occupied in the mental and spiritual life of that time. He is, so

far as it was possible within the limits of the ideas of his time, the

spokesman of free science, the prophet of the newly awakened im-

pulse toward real and independent knowledge. Abelard (and with

him Gilbert) is first of all a rationalist; thought is for him the norm

of truth. Dialectic has the task of distinguishing between true

and false. He may, indeed, subject himself to revelation preserved

in tradition, but, he says, we believe divine revelation only because it

is reasonable. Hence dialectic has, in his case, no longer really the

task which Anselm, following Augustine, prescribed it, of making

the content of faith comprehensible for the intellect ; he demands

for it also the critical right of deciding in doubtful cases according

to its own rules. Thus, in the treatise "Sic et iVbn," he set the

views of the Church Fathers over against each other to their recip-

rocal disintegration dialectically, in order to find at last what is

worthy of belief only in what is capable of proof. So, too, in his

Dialogas, the cognising reason appears as judge over the various

religions, and while Abelard regards Christianity as the ideal con-

summation of the history of religions, there are expressions in his

works 1 in which he reduces the content of Christianity to the origi-

nal moral law, which was re-established by Jesus in its purity.

From this standpoint, too, Abelard was the first to win once more

a free, unbiassed view for the interpretation of antiquity. Little

as he knew of them, he was an admirer of the Greeks ; he sees in

their philosophers Christs before Christianity, and regarding menlike Socrates and Plato as inspired, he asks (reversing the thought

of the Church Fathers, cf. p. 223, note 5) whether religious tradi-

tion may not perhaps have been partly created by these philoso-

phers. Christianity is regarded by him as the philosophy of the

Greeks made democratic.

Abelard, like almost all the " Enlighteners " of the Middle Ages,2

was an obedient son of the Church. But if this fact were to put us

in error as to the significance of his personality in the line just

mentioned, — a significance rather for the history of religion and

civilisation than as producing something philosophically new, — it

would be sufficient to take into account the attacks which he met.

1 Cf. the evidence for what follows in Eeuter, Gesch. der Aufklärung imM.-A., I. 183 ff.

2 A. Harnack, Dogmengeschichte, III. 322.

Page 321: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 24.] Dualism of Body and Soul. 301

In fact, his controversy with Bernard of Clairvaux is the conflict of

knowledge with faith, of reason with authority, of science with the

Church. And if Abelard lacked ultimately the weight and staying

power of personality to prevail in such a contest, 1it will be remem-

bered, on the other hand, that a science such as the twelfth century

could offer— even aside from the external power to which the

Church at that time had attained— must have been inferior to the

mighty inward strength of faith, even if it had not been supported

by so great and high a personality. For that bold postulate, so full

of the future, that only unprejudiced scientific insight should deter-

mine faith, — what means did it then possess for its fulfilment ?

Its only means were the hollow rules of dialectic ; and the content

which this science had to exhibit, it owed just to that tradition

against which it rebelled with its intellectualistic criticism. This

science lacked the material strength to carry out the part to whichshe felt herself called ; but she set herself a problem which, while

she herself was not able to solve it, has never again vanished from

the memory of European peoples.

We hear, indeed, of the disturbing practices of those who would

have everything treated only " scientifically "

;

2 complaints multiply

after the time of Anselm over the growing rationalism of the Zeitgeist,

over the evil men who will believe only what they can comprehend

and prove, over the Sophists who, with impudent dexterity, knowhow to dispute pro et contra, over the " deniers," who from ration-

alists are said to have become materialists and nihilists ;— but not

even the names of the men who answer to this description have

been preserved, to say nothing of their doctrines. And just this

lack in proper material of its own was the reason that the dialectic

movement, whose prince was Abelard, in spite of all its zeal and all

its acuteness, ran out and became exhausted without direct and

immediate results.

§ 24. The Dualism of Body and Soul.

On these grounds it is explicable that in the twelfth and, in part,

even in the eleventh century, we find the feeling of the unfruitful-

ness of dialectic as widely extended as the feverish impulse to

attain through it to true knowledge. A tendency that indicates

disillusion is manifested in this period by the side of the ardent

desire for knowledge. Discontented with the subtilties of dialectic,

which, even in men like Anselm, had laid itself under obligation to

1 Cf. Th. Ziegler, Abaelarcfs Ethica, in Strassburg. Abh. z. Philos. (Freiburg,

1884), p. 221.12 "PvH philosophi."

Page 322: Windelband, History of Philosophy, Vol. 1

302 Mediceval Philosophy : First Period. [Part IIL

place the ultimate mysteries of faith upon a rational basis, some

plunged from unfruitful theory into practical life, " in das Rauschen

der Zeit, ins Rollen der Begebenheit,"— into the rush of time, the

rolling of events,— others plunged into a revelry in supra-rational

Mysticism ; others, finally, into diligent work in empirical research.

All the opposites, into which an intellectual activity that is predom-

inantly logical can pass over, develop by the side of dialectic, and

take their position against it in a more or less firmly concluded

league, — Practice, Mysticism, and Empiricism.

There resulted from this at first a peculiarly distorted relation to

scientific tradition. Aristotle was known only as the father of

formal logic and master of dialectic, and in consequence of this igno-

rance was regarded as the hero of the purely intellectual mode of

considering the world. Plato, on the contrary, was known partly

as the creator of the doctrine of Ideas (unwittingly falsified in

accordance with Neo-Platonic processes), partly, by virtue of the

preservation of the Timceus, as the founder of a philosophy of

Nature whose fundamental teleological character found the live-

liest assent in religious thought. Hence when Gerbert, as a counter-

poise against the pride of dialectic in which he himself had at first

made some not very successful attempts, commended the study of

Nature, to which he had been stimulated by the example of the

Arabians, and which corresponded to his own vigorous practical

bent toward active life, he could count on approval for this en-

deavour only among men who, like him, were working toward an

extension of material information, and who, in aid of this, were

appropriating the results of ancient researches. Thus the return to

antiquity makes here its first appearance as the source of material

knowledge in opposition to the Aristotelian dialectic, — a first weakRenaissance which, half humanistic, half naturalistic, aims to gain

a living content of knowledge. 1 Gerbert's disciple, Fulbert (died

1029), opened the school of Chartres, which, in the following period,

became the seat of the Platonism that was intimately associated

with the study of Nature. Here worked the brothers Theodoric

and Bernard of Chartres ; from this school William of Conches

received his tendency. In their writings the powerful stimulus of

classical antiquity unites with the interest of an active and vigorous

1 The cloister Monte Cassino in Italy formed one of the main seats of this

movement. Here (about 1050) the monk Constantinus Africanus worked, who,as is known to have been the case also with the Platonist Adelard of Bath,gathered his learning on his journeys in the Orient, and was especially active

in the translation of medical treatises by Hippocrates and Ga'Jen. The effects

of the activity in this cloister are shown not only in literature, but also in thefounding of the famous school of Salerno in the middle of the twelfth century.

Page 323: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 24.] Body and Soul : School of Chartres. 303

knowledge of Nature. We see here one of the most peculiar shift-

ings that have occurred in the history of literature. Plato and

Aristotle have exchanged their roles : the latter appears as the ideal

of an abstract science of conceptions, the former as the starting-

point for a concrete knowledge of Nature. The knowledge of ex-

ternal reality that meets us in this period of mediaeval science is

attached to the name of Plato. So far as there is a natural science

in this age, it is that of the Platonists, — of a Bernard of Chartres,

of a William of Conches, and their associates. 1

But this disposition toward concrete reality, which makes the

Platonists of the Middle Ages conspicuous as contrasted with the

high-soaring metaphysics of the dialecticians, assumed still another

form, which was much more valuable. Incapable as yet of gaining

from outer experience better results than those already at its hand

in the transmitted Greek science, the empirical impulse of the

Middle Ages directed its activity to the investigation of the mental

life, and unfolded the full energy of real observation and acute

analysis in the domain of inner experience— in psychology. This is

the field of scientific work in which the Middle Ages attained the

most valuable results.2 In this, the experience of practical life as

well as that of the sublimest piety was filled with a substantial con-

tent, and as such set itself in opposition to the dialectical play of

conceptions.

1. The natural leader in this field was Augustine, whose psychologi-

cal views exercised a mastery that was the stronger in proportion as

his views were interwoven with the current religious conviction, and

in proportion, also, to the slight extent to which the Aristotelian

psychology was known. But Augustine had maintained in his

system the complete dualism which regarded the soul as an imma-

teri?l substance, and man as a union of two substances, body and

soul. Just for this reason he could not expect to gain a knowledge

of the soul from its relations to the body, and took with full con-

sciousness of his procedure the standpoint of inner experience.

The new principle of method which had thus arisen from meta-

physical presuppositions could unfold itself undisturbed so long as

the monistic metaphysical psychology of the Peripatetic school re-

1 This humanistic natural science of the early Middle Ages was not at all

discriminating in its adoption of ancient tradition ; so, for example, if we maytrust the account of Walter of St. Victor (in the extracts made by Bukeus,

Migne, Vol. 190, p. 1170), William of Conches regarded an atomistic conception

of Nature as capable of union with his Platonism. (Migne, Vol. 90, pp. 1132 it.).

2 Cf. for this and for what follows (as also for § 27, later) the articles byH. Siebeck in Vols. I.— III. of the Archiv für Geschichte der Philosophie, andalso in Vols. 93, 94, Zeitschrift für Philos. u. philos. Krit. (1888-90).

Page 324: Windelband, History of Philosophy, Vol. 1

304 Mediceval Philosophy : First Period. [Pakt IIL

maiued unknown. And this unfolding was furthered emphatically

by those needs which brought the Middle Ages to psychology.

Faith sought knowledge of the soul for the purpose of the soul's

salvation, and this salvation was found just in those transcendent

activities through which the soul, estranged from the body, strives

toward a higher world. It was, therefore, principally the Mystics

who sought to spy out the secrets of the inner life, and thus became

psychologists.

Weightier and philosophically more significant than the individual

doctrines propounded in this line, which were often very fantastic

and hazy, is the fact that by means of these and connected theories,

the dualism of the sensuous and super-sensuous worlds was maintained

in its full strength, and thus formed a strong counterpoise to the

Neo-Platonic monism. But it was not destined to exercise this

metaphysical influence till later : at first, in the more limited form

of the anthropological dualism of body and soul, it became the

starting-point for psychology as the science of inner experience. 1

It is, therefore, a very noteworthy phenomenon that the sup-

porters of this psychology as " natural science of the inner sense,"

as it was later called, are precisely the same men who are faithfully

exerting themselves to gain a knowledge of the outer world from all

available material. Having turned away from dialectic, they seek a

knowledge of what is real in experience, a philosophy of Nature

;

but they divide this into two completely separated fields, physica

corporis and physica animce. Among the Platonists the preference

for the study of external Nature is predominant, among the Mystics

that for the study of the internal Nature. 2

2. But we must regard as the characteristic, the essentially newand beneficial mark of this empirical psychology, the endeavour,

not only to classify the psychical activities and states, but to appre-

hend them in the living stream of mental life, and to comprehend

their development. These men in their pious feelings, in their

struggles for the enjoyment of divine grace, were conscious of an

inner experience, of a history of the soul, and were impelled to write

this history ; and while in so doing they used Platonic, Augustinian,

1 Cf. also K. Werner, Kosmologie und Naturlehre des scholastischen Mit-telalters, mit specieller Beziehung auf Wilhelm von Conches ; and Der Entwick-lungsgang der mittelalterlichen Psychologie von Alcuin bis Albertus Magnus(off-prints from the Sitzungsberichten (Vol. 75), and Denkschriften (Vol. 25)respectively of the Vienna Acad., 1876).

3 Nevertheless it must be mentioned that Hugo of St. Victor not only showsan encyclopaedic knowledge in his Eruditio Didascalica, but also shows that heis acquainted, even to the most exact detail, with the teachings of ancient medi-cine, particularly with the theories of physiological psychology (explanation of

perceptions, temperaments, etc.).

Page 325: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 24.] Body and Soul : Victorines. 305

and Neo-Platonic conceptions in motley mixture to designate in-

dividual facts, the essential and decisive point is that they under-

took to exhibit the development of the inner life.

These Mystics, who were not seeking a metaphysics but already

possessed one in their faith, were not much troubled by the ques-

tion which later became so important, of how this duality of body

and soul should be understood. Hugo of St. Victor is indeed con-

scious that though the soul is lowest in the immaterial world, and

the human body highest in the material world, the two are yet so

opposite in constitution that their union (unio) remains an incom-

prehensible enigma ; but he thinks that in this very fact God has

shown, and desired to show, that for him nothing is impossible.

Instead of racking their brains dialetically upon this point, the

Mystics rather assume this dualism as a presupposition, in order to

isolate the soul for their scientific consideration, and to observe its

inner life.

This life, however, is, for Mysticism, a development of the soul to

God, and so this firstform of the psychology of the inner sense is the his-

tory of salvation in the individual soid. The Mystics regarded the soul

essentially as Gemilth [" heart," the seat of sentiment and feeling,

rather than intellect]. They show the development of its vital pro-

cess out of the feelings, and prove their literary virtuosoship in their

depicting of the states and movements of feeling. They are also

the genuine successors of Augustine in examining, in their analysis

of this process, the motive forces of the will, in investigating the

decisions of the will, by virtue of which faith conditions the course

of knowledge, and finally in the fact that they ultimately regard as

the highest stage in the soul's development the mystical contempla-

tion of God, which, to be sure, is here held to be the same with love.

Such, at least, was the activity of the two Victorines, Hugo and

Richard, who were completely sustained by the spirit of science,

while in the case of Bernard of Clairvaux, the practical factor of the

will is much more strongly emphasised. Bernard is unwearied in

denouncing as heathenish that pure impulse after knowledge for its

own sake which comports with all the virtues and vices, and yet,

even for him, the last of the twelve stages of humility is that

ecstasy of deification with which the individual disappears in the

eternal essence, " as the drop of water in a cask of wine."

The psychology of knowledge, also, is built up with the Victorines

upon Augustinian lines. Three eyes are given to man, — the eye of

flesh to know the corporeal world, the eye of reason to know himself

in his inner nature, the eye of contemplation to know the spiritual

world and the deity. While, then, according to Hugo, cogUatio,

Page 326: Windelband, History of Philosophy, Vol. 1

306 Mediaeval Philosophy : First Period. [Part III

meditatio, and contemplatio are the three stages of intellectual activ-

ity, the degree to which he emphasises the co-operation of the imag-

ination (imaginatio) in all kinds of knowledge is interesting and

characteristic of his personality. Even contemplation is a visio

intellectualis, a mental beholding which alone grasps the highest

truth undistorted, while thought is not capable of this.

Old and new are thus variously mingled in the writings of the

Victorines. Fantasies of mystic rapture force their way amid the

most acute observations and the most delicate portrayals of the psy-

chical functions. The method of self-observation doubtless falls here,

too, into the danger of leading to Schwärmerei, 1 or ecstatic enthusi-

asm ; but, on the other hand, it wins much fruit of its own, it

breaks up the soil for the research of the future, and, above all, it

marks off the field on which modern psychology is to grow.

3. This new science received support and enrichment likewise

from quite another direction : a side-result of the controversy over

universals— and that, too, not the worst result— came to its aid.

When Nominalism and Conceptualism combated the doctrine that

universals exist in themselves, and declared the species and genera

to be subjective creations in the knowing mind, the duty fell on

them of making intelligible the process by which these universal

ideas arise in the human mind. They found themselves thus sent

directly to the empirical study of the development of ideas, and sup-

plemented the sublime poesy of the Mystics with results which were

indeed sober and dry, but all the more valuable on that account.

For, just because the matter in hand required an exhibition of the

origin of purely subjective contents of thought, which were to be

explained as the products of man's development in time, this inves-

tigation could become only a contribution to the psychology of inner

experience.

The very thesis of extreme Nominalism afforded its opponents

occasion to treat the relation of word to thought, and in the case of

Abelard led to a searching investigation of the co-operating activity

that belongs to language in connection with the development of

thought. The question as to the meaning of signs and designations

in the movement of ideas was by this means raised anew. A still

deeper entrance into the heart of theoretical psychology was made

by the investigation which is conducted as to the necessary connec-

tion between intellect and perception in the treatise De Intellectibus.

It is here shown how sensation, as confused idea (confusa conceptid),

enters into the perception (imaginatio) which grasps and holds it

1 Cf. Kant, Anthropologie, § 4.

Page 327: Windelband, History of Philosophy, Vol. 1

Chap, i, § 2-1.] Body and Sold : John of Salisbury. 307

together with others, and remains preserved reproducible in this

imagination ; how, then, the understanding by successively running

through this manifold material (discursive activity) elaborates it to

concepts and judgments ; and how, after all these conditions have

been fulfilled, opinion, faith, and knowledge arise, in which ulti-

mately the intellect knows its object in a single collective perception

or intuition (intuitive activity).

In a similar way John of Salisbury set forth the process of

psychical development : but in his case the tendency peculiar to the

Augustinian conception of the soul asserts itself most strongly, — the

tendency to regard the different forms of activity not as strata

lying above one another or beside one another, but as ways of

functioning in which the same living unity manifests itself. Hesees already in the sensation, and in a higher degree in perception

or imagination, an act of judgment; and as union of the newly

entering sensations with those which are reproduced, imagination

contains at the same time the emotional states (passiones) of fear

and hope. Thus out of imagination as fundamental psychical state

develops a twofold series of states of consciousness ; in the

theoretical series appear first, opinion, and by comparison of

opinions, knowledge and rational conviction (ratio), both in con-

nection with prudence (prudentia), which is an operation of the

will; finally, by virtue of the striving after calm wisdom (sapientia),

we have the contemplative knowledge of the intellect ;— in the

practical series are given the feelings of pleasure and pain with all

their diversifications in the changing states of life.

Thus with John we have indicated the whole programme of the

later associational psychology in which his countrymen were to

become leaders. And he may be regarded as their prototype not

only in his problems, but also in the mode of their treatment. Hekeeps at a distance from the speculations of dialectic that were so

alien to the active world ; he has the practical ends of knowledge

in his mind, he desires to find his way in the world in which manis to live, and above all in man's actual inner life, and brings with

him into philosophy a fineness and freedom of mind character-

istic of the man of the world, such as aside from him we do not find

at that time. He owes this in no small degree to the education of

the taste and of sound cosmopolitan thought which classical studies

afford ; and in this, too, his countrymen have followed him, not to

their injury. He is the precursor of the English Enlightenment as

Abelard is of the French. 1

1 Reuter, op. tit., II. 80, sets thus Roger Bacon and Abelard over against eachother

;yet precisely the decisive tendency of empirical psychology is present

more strongly in the case of John.

Page 328: Windelband, History of Philosophy, Vol. 1

308 Mediaeval Philosophy : First Period. [Part III

4. We notice finally Abelard's ethics as a peculiar side-phenomenon

in tliis process of making more rigid the contrast of outer and inner,

and of transferring the scientific first principle to the inner nature. 1

Its very title, Scito Te Ipsum, announces it as a science based on

inner experience, and its importance consists just in the fact that

here for the first time ethics is again treated as a proper philo-

sophical discipline, and freed from dogmatic metaphysical efforts.2

This is true of this ethics although it, too, proceeds from the

Christian consciousness of sin as its fundamental fact. But here

it strives to go at once to the heart of the matter. Good and evilr

it says, consist not in the outward act, but in the action's inner-

cause. Nor yet do they consist in the thoughts (suggestio), feelings,

and desires (delectatio) which precede the decision of the will, but

solely in this resolve or consent to the deed (consensus). For the

inclination (voluntas), founded in the whole natural disposition and

in part in the bodily constitution, which may lead toward good or

evil, is not itself in the proper sense good or evil. Fault or error

(vitium)— to this Abelard reduces inherited sin— becomes sin

{peccatum) only through the consensus. But if this is present, the

sin is fully and completely there with it, and the bodily executed

action with its external consequences adds nothing ethically.

The essence of the moral is thus placed by Abelard solely in the

resolve of the will (animi intentio). But what now is the norm

according to which this resolve of the will is to be characterised as

good or evil ? Here, too, Abelard rejects with contempt all external

and objective determination by a law ; he finds the norm of judg-

ment solely within the deciding individual, and it consists in the

agreement or non-agreement with the conscience (conscientia) . That

action is good which is in accord with the agent's own conviction

;

that only is bad which contradicts this.

And what is conscience ? Where Abelard teaches as a philoso-

pher, as the rationalistic dialectician that he was, there conscience

is for him (in accordance with ancient example, Cicero) the natural

moral law, which, though known in varying degree, is common to

all men, and which, as Abelard was convinced, was wakened

to new clearness in the Christian religion, after it had become ob-

scured through human sin and weakness (cf. above, § 23, 7). But

1 Cf. on this Th. Ziegler in the Strassburger Abhdl. z. Phil. (Freiburg,

1884).2 It throws a surprising light upon the clearness of Abelard' s th ought when

he incidentally separates the metaphysical conception of the good (perfection =reality) carefully from the moral conception of the good, with which alone ethics

has to do. He shows in this that he had penetrated this complication of prob-

lems, one of the most intricate in history.

Page 329: Windelband, History of Philosophy, Vol. 1

Chap. 1, § 24.] Body and Soul: Abelard. 309

for the theologian this lex naturalis is identical with the will of God. 1

To follow the conscience means, therefore, to obey God; to act against

the conscience is to despise God. But where the import of the

natural moral law is in any wise doubtful, the only resort for the

individual is to decide according to his conscience, that is, according

to his knowledge of the divine command.

The ethics of intention 2 which was presented by the head of the

dialecticians and Peripatetics proves itself to be an enhancement of

the Augustinian principles of internalisation and of the individual-

ism of the will, which forces its way out of the system of the great

Church teacher and beyond its bounds, to fruitful operation in the

future.

1 In his theological metaphysics Ahelard seems occasionally to have gone sofar as to reduce the content of the moral law to the arbitrary choice of thedivine will {Commentary on the Epistle to the Bomans, II. 241).

2 The important contrast here presented in various directions to Churchtheory and practice cannot be brought out here.

Page 330: Windelband, History of Philosophy, Vol. 1

CHAPTER II. SECOND PERIOD.

(After about 1200.)

Karl Werner, Der hl. Thomas von Aquino. 3 vols., Regensburg, 1858 ff.

Karl Werner, Die Scholastik des späteren Mittelalters. 3 vols., Vienna, 1881 ff.

The felt need for real knowledge, which mastered Western science

after the first enthusiasm for dialectic was past, was very soon to

find a satisfaction of unsuspected extent. Contact with the Oriental

civilisation which at first maintained itself victoriously against the

shock of the Crusades, disclosed to the peoples of Europe new worlds

of intellectual life. Arabian, and in its train Jewish, science ] madetheir entry into Paris. They had preserved the tradition of Greek

thought and knowledge more immediately and more completely than

had the cloisters of the West. A stronger and richer stream of

scientific material poured over Bagdad and Cordova than over Romeand York. But the former brought not much more that was newwith it than did the latter. Rather, as regards thoughts which dis-

cover or establish principles, the Oriental philosophy of the Middle

Ages is still poorer than the European. Only, in the breadth and

quantity of tradition, in the compass of learned material and in

the extent of information in matters of science, the East was far

superior, and these treasures now passed over into the possession of

the Christian peoples.

From the point of view of philosophy, however, the matter of

chief importance was that Parisian science became acquainted not

1 The author believes that he may and ought to decline to give a full exposi-tion of the Arabian and Jewish philosophy of the Middle Ages— ought to, in sofar as he is here in great part excluded from penetrating to the original sources,

and would therefore find himself forced to reproduce others' expositions at

.second hand,— may, however, because that which passed over with fructifying

influence into European science from this large literature— and it is only this

element that could be treated in this presentation of the development of philos-

ophy as a whole — is found to be, with very small exceptions, the spiritual

possession of antiquity, of the Greek or the Hellenistic philosophy. On this

account there will be given only a brief survey of the Arabian and Jewish phi-

losophy in the Middle Ages, which will be found at the close of the introductorymaterial of this chapter, pp. 310-318.

310

Page 331: Windelband, History of Philosophy, Vol. 1

Chap. 2.] Second Period. 311

only with the entire logic of Aristotle, but also with all parts of his

philosophy that furnished material knowledge. By this " newlogic " fresh blood was infused into the already dying dialectic, and

while the task of rationally expounding the view of the world held

by faith was attacked anew and with a matured technique of thought,

there was presented at the same time an almost immeasurable mate-

rial for arrangement in the metaphysico-religious system.

Mediaeval thought showed itself abundantly ready for the problem

thus enhanced, and solved it under the after-working of the impres-

sion of that most brilliant period in the development of the papacy

which Innocent III. had brought about. The Neo-Platonic-Arabian

Aristotelianism, which at the first, with its naturalistic consequences,

seemed only to strengthen the rationalistic courage of dialectic to

victorious pride, was mastered with admirable swiftness and bent to

the service of the system of the Church. This, indeed, was possible

only in a form in which the intellectualistic elements of Augustinian

thought and those allied to Neo-Platonism gained a decided pre-

ponderance in this now completely systematic development of a

philosophy conformed to the doctrine of faith. In this way was

completed an adjustment and arrangement of world-moving thoughts

upon the largest and most imposing scale that history has seen,

and that, too, without the creative activity of any properly newphilosophical principle as its impulse toward the formation of a

system. The intellectual founder of this system was Albert of Boll-

stcidt. It owes its organic completion in all directions, its literary

codification, and thus its historical designation, to Thomas Aquinas,

and finds its poetical exposition in Dante's Divine Comedy.

But while Hellenistic science and Christian faith seemed to be

brought into complete harmony in Thomism, the opposition between

them broke forth at once all the more violently. Under the influ-

ence of Arabian doctrines, the pantheism involved in the logical

consequence of Realism from being potential became actual in ex-

tended circles, and immediately after Thomas, his fellow-Domin-

ican, Master Eckhart, developed scholastic intellectualism to the

heterodoxy of an idealistic Mysticism.

Hence it is comprehensible that Thomism also encountered the

resistance of a Platonic-Augustinian tendency, which indeed gladly

adopted the increase in the knowledge of Nature (as had been the

case before) and the perfection of the logical apparatus, but putaside the intellectualistic metaphysics and developed all the moreenergetically the opposite elements of Augustinianism.

This tendency reached its full strength in the acutest and deepestthinker of the Christian Middle Ages, Duns Scotus, who brought the

Page 332: Windelband, History of Philosophy, Vol. 1

312 Mediaeval Philosophy. [Part III.

germs of the philosophy of the will, contained in Augustine's

system, to their first important development, and so from the meta.

physical side gave the impulse for a complete change in the direc-

tion of philosophical thought. With him religious and scientific

interests, whose fusion had begun in the Hellenistic philosophy,

begin to separate.

The renewal of Nominalism, in which the intellectual movement

of the last century of the Middle Ages culminated in an extremely

interesting combination, led to the same result with still more last-

ing force. Dialectic, which had anew obtained the mastery and

was flaunting itself in various disputations, developed in its text-

books on logic the Aristotelian schematism. This was worked out

especially on the grammatical side, and there developed to a theory

which attached the doctrine of judgment and the syllogism to the

view that regarded the concepts (termini) as subjective signs for

really existing individual things. This Terminism became united

in William of Occam with the naturalistic tendencies of the Arabian-

Aristotelian theory of knowledge, and these combined combated

Realism, which had been maintained alike in Thomism and Scotism.

But Terminism also became united with the Augustinian doctrine

of the will into a powerful individualism, with the beginnings of

the empirical psychology which studied the history of develop-

ment, to a kind of idealism of the inner experience, and with the

natural investigation which was conquering wider and wider territory,

to an empiricism that was to be fruitful in the future. Thus under

the scholastic covering were sprouting the germs of new thought.

Here and there in this extremely diversified movement men still

vainly appear with the confidence that they can create a rational

system of religious metaphysics, and finally a man of the signifi-

cance of Nicolaus Cusanus sought vainly to force all these elements

of a new secular science back under the power of a half scholastic,

half mystic intellectualism : it was just from his system that those

elements exercised an influence upon the future, that was all the

stronger because of his work.

The reception of Aristotle falls in the century 1150-1250 (for this topic see

principally the work of A. Jourdain, cited p. 273). It began with the more val-

uable parts of the Organon, hitherto unknown (yetus— nova logica), and pro-

ceeded to the metaphysical, physical, and ethical books, always accompaniedby the introduction of the Arabian explanatory writings. The Church slowlyadmitted the new logic, although dialectic was again set in fluctuation thereby

;

for it soon became convinced that the new method which was introduced withthe aid of the doctrine of the syllogism, was advantageous for presenting its

own teachings.

This scholastic method in the proper sense is as follows : a text used as

the basis for discussion is broken up by division and explanation into a numberof propositions

;questions are attached and the possible answers brought to-

Page 333: Windelband, History of Philosophy, Vol. 1

£hap. 2.] Second Period. 313

gether ; finally the arguments to be adduced for establishing or refuting these

answers are presented in the form of a chain of syllogistic reasoning, leading

ultimately to a decision upon the subject.

This scheme was first employed by Alexander of Halss (died 1245) in his

Summa Universes Theologioe, with a mastery which was far superior to the

mode of treatment of the earlier Summists in wealth of contents, clearness of

development, and definiteness of results, and was scarcely surpassed even later.

An analogous change in method was worked out with regard to the material

in the encyclopaedias of natural science by Vincent of BeauvaiB (VincentiusBellovacensis, died about 1265), by his Speculum Quadruples, and JohannesFidanza, called Bonaventura (1221-1274), did the same work for the doctrines

of Mysticism, especially those of the Victorines. Among Bonaventura^ worksthe Beductio Artium ad Theologiam is especially characteristic. Cf. K. Werner,Die Psychologie und Erkenntnisslehre des B. (Vienna, 1876).

The Church proceeded in a much more hesitating manner in regard to Aris-

totle's Metaphysics and Physics, because these made their entrance in intimateconnection with Averroism, and because this latter theory had developed to

open pantheism the Neo-Platonic Mysticism which had never been entirely

forgotten since Scotus Erigena. As the defenders of such a system appearAmalrich of Bena near Chartres, and David of Dinant, about 1200, concern-ing whose doctrines we are informed only by later writers, especially Albertand Thomas. With the widely extended sect of the Amalricans, which, after

the Lateran council of 1215, was persecuted with fire and sword, the ''EternalGospeV of Joachim Floris was also connected. Cf. on this J. N. Schneider(Dillingen, 1873).

The judgment of condemnation passed upon the Averroistic Pan-psychism(cf . § 27) applied at first to Aristotle also. It is the service of the two men-dicant orders, the Dominicans and Franciscans, to have broken this connec-tion, and to have brought over the power of the Church to the recognition of

the Peripatetic system. By a long conflict, which frequently wavered this wayand that, they succeeded in founding two chairs of the Aristotelian philosophyat the University of Paris, and finally in having them taken into the faculty

(cf. Kaufmann, Gesch. d. Univ., I. 275 ff.). After this victory in 1254, respect

for Aristotle rose fast, until he became the highest philosophical authority. Hewas praised as the forerunner of Christ in matters of Nature as was Johnthe Baptist in matters of grace, and from this time on Christian science (like

Averroes) held him to be in such a sense the incarnation of scientific truth, that

in the following literature he is often cited only as " Philosophus."The doctrine of the Dominicans, which has remained until the present time

the official doctrine of the Catholic Church, was created by Albert and Thomas.Albert of Bollstädt (Albertus Magnus) was born 1193 at Lauingen in

Swabia, studied in Padua and Bologna, taught in Cologne and Paris, becameBishop of Regensburg, and died in Cologne in 1280. His writings consist for

the most part of paraphrases and commentaries upon Aristotle ; aside from the

Summa his Botany is particularly of independent value (De Vegetabilibus.

Libri VII. ; ed. by Meyer and Jessen, Berlin, 1867). Cf. J. Sighart, AI. Mag.sein Leben und seine Wissenschaft (Eegensburg, 1857) ; v. Hertling, AI. Mag.und die Wissenschaft seiner Zeit (hi Hist. -pol. Blätter, 1874) ; J. Bach, Al.

Mag. (Vienna, 1888).Thomas of Aquino, born 1225 or 27 in Roccasicca, Lower Italy, was edu-

cated at first in the cloister Monte Cassino, famous of old for study in natural

science, then in Naples, Cologne, and Paris. After this he taught alternately

at these universities and also at Rome and Bologna, and died, 1274, in a cloister

near Terracina. Besides minor treatises, his works contain commentaries onAristotle, on the Liber de Causis and the Sentences of Peter Lombard, andin addition to these, principally the Summa Theologice and the treatise Deveritate fidei Catholicce contra gentiles (Summa contra gentiles). The treatise

De Begimine Principum belongs to him only in part. From the very copiousliterature concerning him, the following may be named: Ch. Jourdain, LaPhilosophie de St. Th. (Paris, 1858); Z. Gonzalez, Studien über die Philos.

des. hl. Th. v. A., translated from the Spanish by Nolte (Regensburg, 1885);It. Eucken, Die Philos. d. Th. v. A. und die Cultus der Nexizeit (Halle, 1886);A. Frohschammer, Die Philosophie des Th. v. A. (Leips. 1889).

Page 334: Windelband, History of Philosophy, Vol. 1

314 Mediceval Philosophy. [Part lei.

The philosophical importance of Dante Alighieri has been best recognised

among his editors by Philalethes in the commentary on his translation of the

Divina Commedia. Besides his great world-poem, the treatise De Monorchiashould not be forgotten in a philosophical consideration. Cf. A. F. Uzanam,D. et la Philosophie Catholique au 13ne Siede (Paris, 1845); G. Baur, Boethius

unci Dante (Leips. 1873).Interest in other Thomists, whose number is great, is only literary-historical.

To the Dominican Order belonged also the father of German Mysticism,Master Eckhart, a younger contemporary of Thomas. Born in the middle of

the thirteenth century, probably in Saxony, at about 1300 he was Professor of

Philosophy in Paris, became then Provincial of his Order for Saxony, lived for

a time in Cologne and Strassburg, and died during the painful discussions con-

cerning the orthodoxy of his doctrine in 1329. The extant writings (collected

by F. Pfeiffer, II. Leips. 1857) are principally sermons, tracts, and aphorisms.Cf. C. Ullman, Beformatoren vor der Beformation, Vol. II. (Hamburg, 1842);W. Preger, Gesch. d. deutschen Mystik im Mittelalter (Leips. 1875, 1881) ; also

the different editions and articles by S. Denifle. On Eckhart in particular,

•J. Bach, M. E. der Vater der deutschen Speculation (Vienna, 1864); A. Lasson,

M. E. der Mystiker (Berlin, 1868).

In its farther development German Mysticism branched into the heresies of

the Beghards and of the " Friends of God " of Basle ; in the case of the former it

led to the most radical connection with the Averroistic pantheism. It took the

form of popular preaching with John Tauler at Strassburg (1300-1361), andof poetic song with Heinrich Suso of Constance (1300-1365) . Its theoretical

doctrines maintained themselves, while the heterodoxy was diminished, in the" German Theology" (first edited by Luther, 1516).

The Augustinian Platonic opposition against the suspected Aristotelianism

of the Arabians has as its main supporters :—

"William of Auvergne, from Aurillac, teacher and Bishop in Paris, where he

died in 1249, author of a work De Universo. He is treated by K. Werner, DiePhilosophie des W. v. A. (Vienna, 1873).

Henry of Ghent (Henricus Gandavensis, Heinrich Gcethals of Muda near

Ghent, 1217-1293), the valiant defender of the primacy of the will against

Thomism. Besides a theological compendium, he wrote a Summa Qucestionum

Ordinarium, and principally Quodlibeta Theologica. Cf . K. Werner, H. v. G. als

Bepräsentant des christlichen Piatonismus im 13 Jahrhundert (Vienna, 1878).

Richard of Middletown (R. de Mediavia, died 1300) and William de la

Marre, the author of a violent Correctorium Fratris Thomce, may also be

named here. In the following centuries an Augustinian theology proper main-tained itself by the side of Thomism and Scotism. iEgydius of Colonna is

regarded as its leader (iEg. Romanus, 1247-1316). Cf. K. Werner, Schol. d.

spät. M.-A., III.

The sharpest opposition to Thomism grew out of the Franciscan order.

Roger Bacon's was a mind fruitfully stimulating in all directions, but not

appearing in a fixed and definite form in any one of them. He was born in

1214, near Ilchester, educated in Oxford and Paris, several times persecuted onaccount of his occupations and theories, which were directed in the line of

natural research, protected only for a time by Pope Clement IV., and died soonafter 1292. His doctrines are embodied in the Opus 31ajus (ed. by Bridges,

Oxford, 1897), and in the form of extracts in his Opus Minus (ed. by Brewer,

Lond. 1859). Cf. E. Charles, B. B., sa vie, ses ouvrages, ses doctrines (Paris,

1861), and K. Werner, in two articles on his psychology, theory of knowledge,

and physics (Vienna, 1879).

The most important thinker of the Christian Middle Ages was JohannesDuns Scotus. His home (Ireland or Northumberland) and the year of his

birth, which was about 1270, are not certainly known. At first a scholar and

teacher in Oxford, he then won high reputation at Paris, where he was active

after 1304, and in 1308 moved to Cologne, where he died soon after his arrival

— all too early. The edition of his works prepared by his Order (12 vols.,

Lyons, 1639) contains, besides the genuine writings, much that is not genuine

or that has been worked over, and especially transcripts of his disputations and

Page 335: Windelband, History of Philosophy, Vol. 1

Chap. 2.] Second Period. 315

lectures. To the latter belongs the so-called Opus Parisiense, which forms a com-mentary upon the Sentences of the Lombard. The Questioned Quodlibetales havea similar origin. The Ojnts Oxoniense, the original commentary upon the Lom-bard, is his own writing. Besides this there are his commentaries upon Aristo-telian writings and some smaller treatises. His doctrine is expounded in Wernerand Stöckl. No exhaustive monograph, corresponding to his importance, exists.

Among his numerous adherents, Francis of Mayro, who died 1325, is the bestknown. The controversy between Thomists and Scotists was a very active oneat the beginning of the fourteenth century, and brought many intermediatetheories into the field ; but soon both parties had to make common cause in

defence against Terminism.Among the logical school books of the later Scholasticism, the most influen-

tial was that of Petrus Hispanus, who died 1277 as Pope John XXI. HisSummuloe Logicales were a translation of a Byzantine-Greek text-book, theSi/i/oi/'ts eis TTjv 'ApitTToriXovs XoytKTjv iivi<jTi)p.rjv by Michael Psellos (in the eleventhcentury). Imitating the processes in this latter treatise (ypdp,aara '4ypa\pe ypa-<pl5i Texffc6s), the well-known barbarous mnemonic designations for the modesof the syllogism were introduced in the Latin version (Barbara, celarent, etc.).

Terminism, developed in the nominalistic direction from this rhetorical andgrammatical logic, contrasted itself as logica moderna with the logica antiquaof the Realists, including both Scotists and Thomists under this latter title.

In the renewal of Nominalism we find "William Durandus of St. Pour-^ain, who died 1332 as Bishop of Meaux, and Petrus Aureolus, who died atParis, 1321, the former coming from Thomism, the latter from Scotism. Muchmore important is William of Occam, the Abelard of the second period. Witha broad and keen vision for reality, and with a bold, unresting eagerness forinnovation, he unites in himself all the elements with the help of which thenew science forced its way out of Scholasticism. Born in a village in theCounty of Surrey, trained under Duns Scotus, he became Professor at Paris,then took an active part in the conflicts of his time between Church and Stateby joining with Philip the Fair and Lewis of Bavaria in combating the papacy,(Disputatio inter clericum et militem super potestate ec^lesiastica prcelatis atqueprincipibus terrarum commissa, and the Defensorium against Pope John XXII.),and died 1347 at Munich. There is no complete edition of his works, but themost important are : Summa Totius Logices, Expositio Aurea super ArtemVeterem, Quodlibeta Septem, Centilogium Theologicum, and a commentary onPeter Lombard. Cf. W. A. Schreiber, Die politischen und religiösen Doctrinenunter Ludwig dem Baier (Landshut, 1858). C. Prantl, Der Universalienstreitim dreizehnten und vierzehnten Jahrhundert (Silz.-Ber. der Münchener Akad.,1874). Occam, too, still waits his philosophically competent biographer.

Of the supporters of terministic Nominalism in the fourteenth century,Johannes Buridan, Rector of the University at Paris, and co-founder of that atVienna, and Marsilius of Inghen, one of the first teachers at Heidelberg, areusually named. A union of mystical doctrines with the nominalistic rejectionof metaphysics is found in Pierre d'Ailly (Petrus de Alliaco, 1350-1425), and inJohannes Gerson (Charlier, 1363-1429).

The attempt at a purely rational exposition of Church doctrine in the interest

of apologetics and propagation was made by Raymundus Lullus of Catalonia(1235-1315), who is principally known by his curious discovery of the "GreatArt," that is, a mechanical device which by combining the fundamental conceptswas intended to present the system of all possible cognitions. An extract fromthis may be found in J. E. Erdmann, History of Phil., I. § 206 [Eng. tr. ed. byHough]. His efforts were repeated in the fifteenth century by Raymund ofSabunde, a Spanish physician, who taught in Toulouse and gained respect by his

Theologia Naturalis (sive Liber Creaturarum) . On him cf. D. Matzke (Breslau,1846); M. Huttler (Augsburg, 1851).

The philosophy of Nicolaus Cusanus (Nicolaus Chrypffs, born in Kues (Cusa)near Trier, 1401, died as Cardinal and Bishop of Biixen, 1464), offers an inter-

esting comprehensive view of the intellectual condition of the departing MiddleAges. The main treatise bears the title De Docta Tgnorantia (ed. in Germantogether with his other most important writings by F. A. Scharpff, Freiburg i. B.1862). Cf. R. Falckenberg, Grundzüge der Philos. des N. v- C. (Breslau. 1880).

Page 336: Windelband, History of Philosophy, Vol. 1

316 Mediaeval Philosophy : Second Period. [Part III

Brief Survey of the Arabian and Jewish Philosophy of the MiddleAges.

This period is certainly more interesting from a literary and historical pointof view than from that of philosophy, and as yet no competent presentation of

the period as a whole has been made. Nor has complete clearness been attainedas yet by investigation, but from the literature concerning it the following areto be emphasised :

—Mohammed al Schahrestani, History of Religious and Philosophical Sects

among the Arabs (German by Haarbriicker, Halle, 1850 f.); A Schmölders,Documenta Philosophice Arabum (Bonn, 1835), and Essai sur les Ecoles Phi-losophiques chez les Ar. (Paris, 1842); Fr. Dieterici, Die Philosophie der Ar. imzehnten Jahrhundert (8 Hefte, Leips. 1865-76). Cf. also Hammer-Purgstall,Gesch. der arabischen Litteratur.

S. Munk, Melanges de philosophic juive et arabe (Paris, 1859), and the sameauthor's articles on the individual philosophers in the Dictionnaire des SciencesPhilosophiques. [W. Wallace, Art. Arabian Phil, in Enc. Brit., Ueberweg,Erdmann. "|

M. Eisler, Vorlesungen über die jüdischen Philosophen des Mittelalters (3vols., Vienna, 1870-84); M.Joel, Beiträge zur Geschichte der Philosophie (Bres-lau, 1876). Cf. also Fürst's Bibliotheca Judaica, and histories of Judaism byGraetz and Geiger.

Close as the relations may be which the philosophy of the two civilised Semiticpeoples sustained to their religious interests, Arabian science especially owesits peculiar character to the circumstance that its founders and supporterswere, for the most part, not members of the clergy, as in the West, but physi-cians (cf. F. Wüstenfeld, Gesch. der arab. Aerzte und Naturforscher, Göttingen,1840). Thus from the beginning the study of ancient medicine and naturalscience went on hand in hand with that of philosophy. Hippocrates and Galenwere as much translated (in part through the medium of the Syrian) and readas were Plato, Aristotle, and the Neo-Platonists. Hence in Arabian metaphysicsdialectic is always balanced by natural philosophy. But well as this was adaptedto afford scientific thought a broader basis of knowledge of facts, we must not,

on the other hand, overestimate the independent achievements of the Arabs in

medicine and natural science. Here, too, mediaeval science is essentially learnedtradition. The knowledge which the Arabs were later able to deliver to theWest had its origin, in the main, in the books of the Greeks. Nor did evenexperimental knowledge experience an essential extension through the Arabs'own work ; only in some fields, as, for example, chemistry and mineralogy andin some parts of medicine, e.g. physiology, do they appear more independent.In their method, however, in their principles by which they apprehend the uni-

verse, and in their entire system of philosophical conceptions, they stand, so far

as our information on the subject reaches, entirely under the combined influenceof Aristotelianism and Neo-Platonism ; and the same is true of the Jews. Norcan it be maintained that a national peculiarity becomes disclosed in their appro-priation of this material. It is rather the case that this whole scientific culturewas artificially grafted upon the Arabian civilisation, it can strike no true rootsinto it, and after a short period of bloom it withers away without vital force.

In the history of science as a whole, its mission is only to give back in part to

the development of the Western mind the continuity which the latter had itself

temporarily lost.

From the nature of the case, the appropriation of ancient science in this casealso was completed gradually and by working backward. Beginning with theNeo-Platonism which was still current in Syrian tradition, and which wasreceived with sympathy on account of its religious colouring, the Arabianthinkers proceeded to ascend to the better sources ; but the consequenceremained that they saw Aristotle and Plato through the spectacles of Plotinusand Proclus. During the rule of the Abassidse an active scientific life prevailedin Bagdad, stimulated especially by the Caliph Almamun at the beginning of

the ninth century. The Neo-Platonists, the better commentators, almost the

entire didactic writings of Aristotle, and the Republic, Laws, and Timceus of

Plato, were known in translations.

Page 337: Windelband, History of Philosophy, Vol. 1

Chap. 2.] Arabian and Jewish Philosophy. 317

The first distinctly emerging personalities, Alkendi, who died about 870, andAlfarabi, who died 950, are scarcely to be distinguished in their teachings fromthe Neo-Platonic elucidators of Aristotle. A greater importance belongs to

Avicenna (Ibn Sina, 980-1037), whose "Canon" became the fundamentalbook of mediaeval medicine in the West, as well as in the East, and who also

exercised a powerful influence by his extremely numerous philosophical writ-

ings, especially his Metaphysics and Logic. His doctrine comes nearer again to

pure Aristotelianism, and perhaps the nearest among all the Arabians.But the extension of these philosophical views was regarded with jealous eyes

by Mohammedan orthodoxy, and the scientific movement experienced so vio-

lent persecutions in the tenth century that it took refuge in the secret league of

the "Pure Brothers." Avicenna himself was also persecuted. The above-named league embodied the extremely excellent compass of the knowledge of

the time in a number of treatises (on this see above, Dieterici), which neverthe-

less, in contrast with Avicenna, seem to show a stronger leaning toward Neo-Platonism.

Of the scientific achievements of their opponents we know on the one handthe strange metaphysics of the orthodox Motekallemin, who, as against the

Aristotelian and Neo-Platonic view of Nature as a living whole, developed anextreme exaggeration of the sole causality of God, and resorted to a distorted

Atomism in the greatest metaphysical embarrassment ; on the other hand, in

the writings of Algazel (1059-1111, Destructio Philosopliorum) there appearsa sceptical and mystical analysis of philosophy.

These latter tendencies won the victory in the Orient the more readily, as the

spiritual exaltation of Mohammedanism quickly declined in that quarter. Thecontinuance of Arabian science is to be sought in Andalusia, where Mohamme-dan civilisation found its short after-bloom. Here, under freer conditions,

philosophy developed to vigorous naturalism, which in turn bore a strongly

Neo-Platonic stamp.A characteristic exposition of the doctrine of knowledge in this philosophy is

found in the Conduct of the Solitary by Avempace, who died 1138, and similar

thoughts culminate with Abubacer (Ibn Tophail, died 1185) in an interesting

comparison of natural with positive religion. The latter author's philosophi-

cal romance The Living One, the Son of the Waking One, which sets forth the

intellectual development of a man upon a lonely island, excluded from all his-

torical and social relations, was published in a Latin translation by Pocock as

Philosophus Autodidactus (Oxford, 1671 and 1700,— not twenty years before the

appearance of Defoe's Bobinson Crusoe!) and in a German translation as

Der Naturmensch by Eichhorn (Berlin, 1783).

But the most important and independent among Arabian thinkers wasAverroes, who was born 1126 in Cordova, was for a time judge, and then

physician in ordinary to the Caliph, was driven afterward by religious perse-

cution to Morocco, and died in 1198. He treated in paraphrases and longer or

shorter commentaries, which were printed in the older editions of Aristotle,

almost all the didactic writings of Aristotle, who was esteemed by him as the

highest teacher of truth. Of his own works (Venice, 1553 ; some exist nowonly in the Hebrew version) the refutation of Algazel, Destructio Destructions,

is most important. Two of his treatises on the relation of philosophy and the-

ology have been published in German translation by M. J. Müller (Munich.

1875). Cf. E. Renan, Averroes et VAverroisme (3d ed., Paris, 1869).

With the expulsion of the Arabians from Spain traces of their philosophical

activity are lost.

Jewish philosophy of the Middle Ages is, in the main, an accompanimentof the Arabian, and dependent upon it. The only exception to this is the Cab-bala, that fantastic secret doctrine whose fundamental outlines, which, to bo

sure, were later much elaborated, show the same peculiar amalgamation of

Oriental mythology with ideas of Hellenistic science as does Christian Gnosti-

cism, and go back to the same period and to the same agitated condition of

thought attendant upon the mingling of religions. Cf. A. Eranck, Systeme de

la Kabbale (Paris, 1842; German by Jellinek, Leips. 1844); II. Joel, Dielieligionsphilosophie des Sohar (Leips. 1849). On the other hand, the mainworks of Jewish philosophy were originally written in Arabic, and not trans-

lated into Hebrew until a relatively late time.

Page 338: Windelband, History of Philosophy, Vol. 1

318 Mediaeval Philosophy : Second Period. [Part III

The book of Saadjah Fajjumi (died 942), Concerning Religions and Philoso-

phies, which aims to furnish an apology for Jewish doctrine, is related to theearliest Arabian Aristotelianism, and still more closely to the free-thinking

Mohammedan theologians, the so-called Mutazilin. In the Neo-Platonic line

we meet Avicebron (Ibn Gebirol, a Spanish Jew of the eleventh century), of

whose Fo7is Vitaz, Hebrew and Latin versions are extant. Moses Maimonidea(1135-1204) is regarded as the most important Jewish philosopher of the MiddleAges. In his culture and doctrine he belongs to the phase of Arabian doctrine

which has Averroes as its centre. His main treatise, Guide to the Perplexed{Doctor Perplexorum) , has been published in Arabic and French with a com-mentary by Munk (3 vols., Paris, 1856-06) [Eng. tr. by Friedlander, Trübner,Lond. ]. The attachment to Averroes is still closer in the case of Gersonides(Levi ben Gerson, 1288-1344).

The Jews, by means of their widely extended mercantile relations, were thechief contributors to the extension of Oriental philosophy in the West, by sale

and translation ; in the thirteenth and fourteenth centuries especially their

schools in Southern France formed the medium for this wide-reaching activity.

To the Arabian and Jewish literature, which was taken up by Christianscience about 1200, belongs finally a number of pseudonymous and anonymouswritings, which arose in the latest periods of Neo-Platonism, and in part per-

haps were of still later date. Among these the principal are the Theology ofAris-totle (Arabic and German by Dieterici, Leips. 1882-83), and the Liber de Causis{De essentia puree bonitatis), an extract from the o-roixeluo-is OeoXoyiK-r) ascribed

to Proclus, published in Arabic, Latin, and German by O. Bardenhewer (Frei-

burg i. B. 1882).

§ 25. The Realm of Nature and the Realm of Grace.

Among all the philosphers of the Middle Ages we find existing,

with greater or less clearness, a lively feeling of the twofold tradi-

tion which forms the presupposition of their thought. In the

earlier period all knowledge and thought had arranged itself, as it

were, of its own accord within the system of religious metaphysics

;

and now there appeared by the side of this a powerful, finely articu-

lated, coherent body of thought which the age, thirsting after real

contents in its barren dialectic, was ready to take up eagerly. The

manifold relations between these two systems which mutually laid

hold upon one another and interpenetrated, determiue the scientific

character of the last centuries of the Middle Ages, and the general

course of the development was, that these antagonistic systems,

starting from an attitude of abrupt opposition, strove toward recon-

ciliation and adjustment, only to diverge all the more violently after

the goal seemed to have been reached. This course of things

appeared as necessarily in the conception of the reciprocal relations

of the different sciences, as in the view of the ultimate relations

of things. In both lines the attempt at synthesis was followed by

a separation that went all the deeper.

The religious thought of the West, whose highest problem had

been to understand the working of divine grace, was confronted by

Oriental philosophy in which the old Grecian philosophical tendency

toward knowledge of Nature had at last attained metaphysical

Page 339: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 25.] The Tivo Realms : Averroism. 319

supremacy : and here, too, again the process of appropriation began

with the adoption of the last consequences, to ascend only by

degrees back to the premises.

1. Hence the form in which Arabian science was first taken up

was that of Averroism. In this, however, science had marked off its

boundaries in the most definite manner as against positive religion.

This had taken place not only in reaction against the attacks to

which the philosophical movement in the East had been subjected,

but still more in consequence of the great mental revolutions which

the age of the Crusades experienced through the intimate contact

of the three monotheistic religions. The more ardently these relig-

ions fought in the sphere of historical reality, the more the sharp-

ness of their contrasting doctrines became blunted from the point

of view of theory. Those who passed through this conflict of relig-

ions as thinking observers could not resist the impulse to seek the

common element behind the differences, and to establish above the

fields of battle the idea of a universal religion. 1 In order to attain

this, every form of special historical revelation must be stripped off,

and the path of universally valid scientific knowledge must be taken.

So with the aid of Neo-Platonic memories, a return was made to the

thought of a universal religion, founded upon science, and the ulti-

mate content of this common conviction was formed by the moral law.

As Abelard in his own way had already reached this result, so

Roger Bacon later, under Arabian influences, designated morality as

the content of the universal religion.

This scientific natural religion, however, had had stamped upon it

more and more by the Arabs the exclusive character of an esoteric

doctrine. The distinction originating with Philo, and current in the

entire patristic thought, between a verbal-historical and a spiritually

timeless sense 2 of religious documents (cf. § 18, 2) here became the

doctrine that positive religion is an indispensable need for the mass

of the people, while the man of science seeks the real truth back of

religion, and seeks it only there,— a doctrine in which Averroes

and Maimonides were at one, and which completely corresponded to

the social relations of Arabian science. For Arabian science always

moved within narrow and closed circles, and as a foreign growth

1 The court of the highly cultured Hohenstaufen Frederick II. in Sicily-

appears as a chief seat of this mode of thought, and in general of the exchangeof thought between East and West.

2 Representing this opinion, the Eternal Gospel of Joachim of Floris wascirculated among the Averroistic Amalricans. This completed for the entire

compass of Christian dogma, the transformation of everything external into theinternal, all the historical into the tunelessly valid : the " pneumatic gospel " of

Origen (cf. § 18, 2) was asserted to have here attained reality, the period of the•spirit" to have begun. Cf. J. N. Schneider (Dillingen, 1874).

Page 340: Windelband, History of Philosophy, Vol. 1

320 Mediceval Philosophy : Second Period. [Part III

never gained true sympathy with the mass of the people: Averroes,

nevertheless, expressly honours Aristotle as the founder of this high-

est, most universal religion of the human race.

Thus in line with this thought, Abubaoer made his " Man in a

State of Nature," who had attained in his isolation to the philosoph-

ical knowledge of God, come into contact again at last with histori-

cal humanity, and in so doing discover that what he had knownclearly and in abstract thought, is here believed in its picturate

wrappings, and that what holds for him as a self-evident demand of

the reason is here extorted from the multitude by means of reward

and punishment.

If now it is hereby admitted that natural and revealed religion

have ultimately the same content, it still follows that they necessa-

rily differ, at least in their expression of the common truth,— that

the conceptions which form the expression of philosophical religion

are not understood by believers, while the picturate ideas of believ-

ers are not regarded as the full truth by philosophers. If, then, by

theology, we understand the exposition of the positive doctrine of

religion, arranged and defended according to the formal laws of

science, i.e. Aristotelian logic,— and this was the form which the

relation of theology to religion had taken in the West as in the

East, — it follows that something may be true theologically which

is not true philosophically, and vice versa. Thus is explained that

doctrine of the twofold truth,1 theological and philosophical, which

went through the entire later Middle Ages, although we cannot

exactly fix the authorship of this formula.2 It is the adequate

expression of the mental state necessarily brought about by the

opposition of the two authorities under which the Middle Ages

stood, viz. Hellenistic science and religious tradition ; and while at

a later time it often served to protect scientific theories from the

persecution of the Church, it was for the most part, even in these

cases, the honest expression of the inner discord in which just the

most important minds of the age found themselves.

2. The science of the Christian peoples accepted this antithesis,

and while the doctrine of the twofold truth was expressly pro-

claimed by bold dialecticians such as Simon of Tournay, or John of

Brescia, and was all the more rigidly condemned by the power of

1 Cf. M. Maywald, Die Lehre von der zweifachen Wahrheit (Berlin, 1871).2 As little can it be fixed with certainty what the origin of that widely ex-

tended formula was, which designated the founders of the three great positive

religions as the three "deceivers" of mankind. Unhistorical, as is every

Enlightenment, the philosophical opposition of that day could explain to itself

only by empirical interests the mythical which could not stan^ before compara-tive criticism.

Page 341: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 25.] The Two Realms : Albert, Thomas. 32J

the Church, the leading minds eould not evade the fact that philos-

ophy, as it had been developed under the influence of Aristotle andthe Arabians, was, and must remain, in its inner nature, alien to

precisely those doctrines of the Christian religion which were spe-

cific and distinctive. With a full consciousness of this opposition,

Albert proceeded to his great task. He understood that the distinc-

tion between natural and revealed religion, which he found in exist-

ence, could no longer be put out of sight, that philosophy and

theology could no longer be identified, but he hoped and laboured

with all his strength that this distinction might not be allowed to

become a contradiction. He abandoned the doctrine that the " mys-

teries " of theology, the doctrines of the Trinity and of the Incar-

nation, can be made rational, and, on the other hand, he corrected in

favour of the Church doctrine the teaching of the " Philosopher

"

on such important points as the question concerning the eternity or

temporal duration of the world. He sought to show that all which

is known in philosophy by the " natural light'' (lumine naturali)

holds good also in theology, but that the human soul can knowcompletely only that, the principles of which it carries within itself,

and that, therefore, in such questions as those in which philosophical

knowledge comes to no finally valid decision and must remain

standing before the antinomy of different possibilities, revelation

gives the decision, — a view in which Albert follows mainly the

results of Maimonides. Faith is meritorious just because it cannot

be proved or established by any natural insight. Revelation is above

reason, but not contrary to reason.

This standpoint for harmonising natural and revealed theology

is essentially that taken by Thomas, although he seeks to limit still

more, if possible, the extent of that which is to be withdrawn from

philosophical insight and given into the possession of faith. Accord-

ing to the fundamental thoughts of his system, moreover, he

apprehends this relation as a relation of different stages of

development, and sees accordingly, in philosophical knowledge, a

possibility given in man's natural endowment, which is brought

to full and entire realisation only by the grace active in revela-

tion.

It is therefore important to notice that Scholasticism, just in this

its highest point, was far from identifying philosophy and theology,

or from making the task of the former, as has often been repre-

sented, an unresting comprehension of dogma. This conception

belongs to the beginnings of mediaeval science, e.g. to Anselm, and

is found sporadically in the times when Scholasticism was entering

upon its dissolution. So, for example, Raymundus Lullus projected

Page 342: Windelband, History of Philosophy, Vol. 1

322 Mediaeval Philosophy : Second Period. [Part Hi.

his "Great Art

"

1 essentially in the opinion that this, by making

possible a systematic explanation of all truths, will be adapted to

convince all " unbelievers " of the truth of the Christian religion.

So, too, later, Raymond of Sabunde aimed to prove with the help of

Lull's Art that if God has revealed himself in a double manner, in

the Bible {liber scriptus) and in Nature (liber vivus), the contents

of these two revelations, of which the one lies at the basis of theol-

ogy, the other at the basis of philosophy, must evidently be the

same. But in the classical time of Scholasticism the distinction

between natural and revealed theology was always kept in mind,

and was drawn the more sharply, the more the Church had occasion

to guard against the confusion of its doctrine with " natural

theology."

3. Hence there were very faithful sons of the Church whobroadened again the cleft between philosophy and theology, and ulti-

mately made it so wide that it could not be bridged. At their head

stands Duns Scotus, who taught that theology should be conceived

and treated only as a practical discipline;philosophy, on the con-

trary, as pure theory. Hence for him and for the co utinuers of his

doctrine, the relation between the two is no longer ühat of supple-

mentation, but that of separation. Between the two opposing terri-

tories of revelation and of rational knowledge, natural theology

shrivels into an extreme poverty of domain. The compass of the

mysteries of theology that are inaccessible for natural knowledge

increases more and more ; with Duns Scotus the beginning of the

created world in time and the immortality of the human soul belong

to this sphere ; and Occam even denies the cogency of the usual

arguments with which rational theology was wont to prove the

existence of God.

This criticism is rooted essentially in the purpose to assure to

faith its just right, and in this purpose it is completely honest. In

connection with the metaphysical dualism which had again become

pronounced (see below, No. 5) the knowledge of the understanding,

bound as it was to sense-perception, seemed incapable of searching

1 This wrong-headed, and yet in many respects interesting and therefore

frequently attempted, discovery, consisted in a system of concentric rings, each

of which bore a group of concepts divided into circular compartments. Byshifting these rings, all possible combinations between concepts were to bebrought about, problems given, and their solutions stated. Thus there was a

Figura A (Dei) which contained the whole theology, a Figura Animse whichcontained psychology, etc. Mnemo-technic attempts, and such as aim at the

discovery of a universal language, or of a system of symbols for expressing

philosophical thoughts, have frequently been attached to this ars combinatoria.

The introduction of the algebraic method of reckoning by letters \s also con-

nected with these efforts.

Page 343: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 25.] The Two Realms : Duns Seotus, Occam. 323

the mysteries of the supernatural world. Thus men like Gerson

based their mystical doctrine precisely upon Nominalism. Thedifference between philosophy and theology is necessary; the con-

tradiction between knowledge and faith is unavoidable. Revelation

has its source in grace, and has the divine realm of grace for its con-

tent; rational knowledge is a natural process of reciprocal inter-

action between the knowing mind and the objects of perception.

Therefore, though Nominalism escaped from the scholastic method

with difficulty, and was late in reaching its goal, it necessarily

ended in regarding Nature as the sole object of science. At all

events, philosophy now set itself as secular science, over against

theology as divine science.

So Duns Scotus and Occam employed language which externally

is quite in harmony with the " twofold truth." That definition of

the boundaries was intended to assert, that in matters of faith dia-

lectic has nothing to say. But it could not fail to be the result,

that in the case of others, this separation would lead to the oppo-

site consequence and back to the original meaning of the claim of

a double truth. It became a charter of liberty for the " secular

philosophy." Dialectical investigation could be pursued even to

the boldest propositions, and yet all offence might be avoided if one

only added that the proposition was so secundum rationem, but that

secundum fidem the opposite was of course true. This occurred so

frequently that the Thomists and Lullists became zealous against it.

In the case of many, to be sure, who availed themselves of this

principle, we cannot doubt that this was their honest opinion ; but

it is just as sure that others, with full consciousness of their pro-

cedure, found in this only a convenient pretext, in order to present

under the protection of this restriction the doctrines of a philosophy

that in its inner spirit was at variance with faith. At all events,

this applies to the school of the Averroists which flourished in

Padua toward the end of the fifteenth century.

4. Parallel to this changeful process of transformation in the

relation between theology and philosophy, and in closest connection

with it, goes an analogous development of metaphysical psychology,

and both have reference in like measure to the fundamental relation

between the supersensuous and the sensuous worlds. Here, too,

dualism is the starting-point, and afterwards again the end. This

dualism had been developed to an especial degree of sharpness by

the Victorines at the close of the first period. In this Mysticism

the last bonds between body and soul were cut, and reconciliation

was made impossible. The spiritual and material worlds fell apart

as separate spheres of the universal reality.

Page 344: Windelband, History of Philosophy, Vol. 1

324 Mediceval Philosophy : Second Period. [Part III.

Now, however, Aristotelianism fulfilled its historical mission of

overcoming the two-worlds theory in Augustine, as formerly in

Plato, and in the Thomist psychology the conception of development,

and of the gradual building up of phenomena, was intended to

bridge that separation. While Hugo of St. Victor had drawn the

dividing line in the created world through the midst of man's nature,

by emphasising the complete impossibility of any comparison be-

tween the two substances there brought together, the human soul

was now to be understood as just that connecting link, through the

medium of which the two worlds come into organic interaction in

the one course of development of all things.

Thomas attains this result by an extraordinarily acute transfor-

mation of the Aristotelian doctrine of Forms and their relation to

matter. The material and the immaterial worlds are characterised

by the fact that, in the latter, pure Forms (formoe. separatee; called

also subsistent Forms) are real or actual as active intelligences with-

out any attachment to matter, while in the former, Forms realise

themselves only in union with matter (inherent Forms). The hu-

man soul, as lowest of the pure intelligences, is a forma separata

(on which rests its immortality) and, at the same time, as entelechy

of the body, it is the highest of those Forms which realise them-

selves in matter. But these two sides of its nature are bound

together in it to an absolute substantial unity, and this unity is the

only Form which is at the same time subsistent and inherent. 1 In

this way the series of individual beings proceeds from the lowest

Forms of material existence, on past plant and animal life, through

the human soul, with uninterrupted continuity over into the world

of pure intelligences— the angels,2 and finally to the absolute Form— the deity. The cleft between the two worlds is closed in Thomism

by this central position of metaphysical psychology.

5. But it seemed to the following period that the cleft was closed

only by being plastered over, as it were, and that the union of so

heterogeneous attributes as the entelechy of the body and the sub-

sistence of a pure intelligence was more of a load than the con-

ception of individual substance was able to bear. Hence Duns

Scotus, whose metaphysics likewise moves naturally within the

Aristotelian terminology, introduced an (inherent) forma corporei-

tatis between the intelligent soul, which he too designates as the

" essential Form " of the body, and the body itself ; and thus the

1 In this is concentrated in a conception the anthropocentric way of viewing

the world, which even Thomism did not overcome.2 Thomas constructs his scale of forms in the material world according to

Aristotle, in the spiritual world according to Dionysius the Areopagite.

Page 345: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 25.] The Two Realms : Thomas, Scotus, Occam. 325

Augustinian and Victorinian separation of the conscious essence

from the physiological vital force was again re-established.

Occam not only made this distinction his own, but, forced to

insert another gradation, analysed the conscious soul into an intel-

lectual and a sensitive part, and ascribed real importance to this

separation. It seems to him that the sensuous activities of con-

sciousness can as little be united with the rational nature whose

vocation it is to behold the immaterial world, as can the form and

motion of the body. Thus for him the soul is split up into a num-

ber of individual faculties, to determine the relation of which

occasions great difficulties, especially with regard to their spatial

inter-relation.

6. The essential thing in this is that the world of conscious-

ness and that of corporeal bodies become again completely sepa-

rated ; and this is shown especially in Occam's theory of knowledge,

which proceeded from these presuppositions to an extremely signifi-

cant innovation.

In their doctrine of the " species inteUigibiles " the two " Realists,"

Thomas and Duns Scotus, had alike followed, though with some vari-

ations, the old Greek idea, that in the knowing process, by means of

the co-operation of the soul and of the external object, a copy of

the latter arises, which is then apprehended and beheld by the soul.

Occam strikes out these species inteUigibiles as a useless doubling 1 of

the external reality, which according to this view, in so far as it is

an object of knowledge, would be assumed as having still another

existence (in psychical reality). But by this act seyisuous knowledge

loses for him its character of being a copy as compared with its object.

An idea {conceptus, intellectio rei) is as such a state or an act of the

soul (passio— intentio animoe), and forms in this a sign (signum)

for the corresponding external thing. But this inner structure is

something of a different nature from the outer reality of which it is

the sign, and therefore it is no copy of it. We can speak of a " re-

semblance " only in so far as in this case the inner reality (esse

objective = content of consciousness) and the outer reality (esse for-

maliter or subjective = objective reality in the present sense of the

word " objective " 2) necessarily relate to each other, and, so to speak,

form corresponding points in the two heterogeneous spheres.

Thus the beginning of a psychological and epistemological idealism

1 According to his methodical principle : entia praeter necessitate™ non esse

multiplicanda.2 The terms "objective" and "subjective" in the Middle Ages have accord-

ingly a meaning exactly the reverse of that which they have in present

usage.

Page 346: Windelband, History of Philosophy, Vol. 1

326 Mediceval Philosophy . Second Period. [Pakt 1IL

develops among the Terminists out of the old duality of mind

and body : the world of consciousness is another world than the

world of things. What is found in the former is not a copy, but

only a sign for something without which corresponds to it. Things

are other than our ideas (ideas) of them.

7. Lastly, Augustine's dualism appeared in its complete bald-

ness in his conception of history. The realm of God and that of the

devil, the Church and the political state, here confronted each other

in rigid antithesis. The historical conditions of which this doctrine

was the reflex, had become changed completely since Augustine's

day. But hitherto the Middle Ages had not only lacked historical

conceptions which would have been adapted to correct this doctrine,

but scientific thought had been employed in such a one-sidedly theo-

logical and dialectical manner, that ethical and social problems had

remained farther outside the horizon of philosophers than had phys-

ical problems. And yet at the same time, history was seeing move-

ments of such grand dimensions that science also must necessarily

take a position with regard to it. If she was able to do this in the

second period in a manner completely worthy of the greatness of

the subject, she owed her strength for this again to the Aristotelian

system, which gave the means into her hand of mastering in thought

the great connected structures of political and historical life, of

arranging in her metaphysics these forms of the series of develop-

ment, and thus of putting into conceptions the mighty import of

that which she was living through. Indeed, in this line in which

the Arabian commentators had not gone before lies the most brilliant

achievement of mediaeval philosophy,1 and since Albert's interest lay

more on the side of physics, the chief credit here falls to Thomas.

Thomas regards the political state, not as did Augustine, as a con-

sequence of the fall, but as a necessary member in the world's life.

In his view, therefore, law or right also flows from the divine nature

and must be so conceived ; above all human institutions stands the

lex naturalis, upon which rest morality and the life of society. In

particular, however, as is proved by language, by the need of help

which the individual feels, and by the impulse toward society, manis by his nature destined for life in a state. The end of the state is,

according to Aristotle's teaching, to realise virtue, and from this end

all the characteristics of the state are to be developed (in philosoph-

ical law— Natural Right or Laiv) . But— and here the new thought

begins — that civic virtue to which the state should educate its

citizens does not exhaust man's destiny. In this he fulfils only his

' Ci W. Dilthey, Einleitung i?i die Geistesivissensrhaften. I. 418 f.

Page 347: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 25.] The Two Realms : Thomas, Dante. 327

purpose as an earthly being ; his higher destiny is the salvation

which grace offers him in the community of the Church. But as

the higher everywhere realises itself through the lower, and the

lower exists for the sake of the higher, the political community is

to be the preparation for that higher community of the State of

God. Thus the state becomes subordinate to the Church as the

means to the end, as the preparatory to the complete. The com-

munity of the earthly life is the school for that of the heavenly—PRJEAMBULA GRATIiE.

By the side of the teleology of Nature which Greek philosophy had

worked out, patristic thought had set the teleology of history (cf. §

21, 6) ; but the two had remained unconnected. The doctrine of the

state set forth by Thomas subordinates the one to the other in a

system of thought, and in so doing completes the most deeply and

widely reaching union of the ancient and Christian conceptions of

the world that has ever been attempted.

With this the capstone is fitted to the metaphysical structure of

Thomism. By this transition from the community of Nature into

that of grace, man fulfils the task which his position in the universe

assigns him, but he fulfils it, not as an individual, but only in the

race. The ancient thought of the state lives again in Christianity

;

but the state is no longer an end in itself, it is the best means for

carrying out the divine world-plan. Gratia naturam non tollit sed

perficit.

8. But even this highest synthesis did not long endure. As in

political life, so also in theory, the relation of Church and state took

on a form that was very much less harmonious. With Dante the

relation of subordination is already exchanged for that of co-ordina-

tion. The poet shares with the metaphysician the thought that

because man's destined end is to be attained only in the race, this

makes a perfect unity in political organisation requisite. Both de-

mand the universal state, the "monarchia" and see in the Empire the

fulfilment of this postulate. But the great Ghibelline cannot think

theocratically, as does the Dominican monk ; and where the latter

assigns to the Imperium the place of subordination beneath the sacer-

dotium, the former sets the two over against each other as powers of

like authority. God has destined man for earthly and for heavenly

happiness in like measure : to the former he is conducted by the

state, by the natural knowledge of philosophy ; to the latter he is

guided by the Church, by means of revelation. In this co-ordination

the joy in the world, characteristic of the Renaissance, bursts forth

as victoriously as does the feeling of strength which belongs to the

secular state.

Page 348: Windelband, History of Philosophy, Vol. 1

328 Mediceval Philosophy : Second Period. [Part III.

And along this line the development proceeded. When the graded

scale of reality constructed by Thomas was severed in the midst of

man's nature, the spiritual and political powers fell apart, as did the

spiritual and corporeal worlds ; and the theory afforded the con-

venient means of banishing the sacerdotium to the supra-mundane

inner nature, and putting the Imperium into sole control within

the world of sense. This is precisely the point of view from which

Occam, in his Disputatio with reference to the controversy between

the papacy and the temporal power, took his position upon the side

of the latter. Nor yet is it any longer possible, in accordance with

his presuppositions, to base the theory of the state upon the realistic

thought of the human race as a whole, bound together for the real-

isation of one end. The Nominalist sees as a substantial back-

ground in social and historical life, only the individuals who will,

and he regards state and society as products of interests (bonum

commune). In theory, as in life, individualism prevails. 1

§ 26. The Primacy of the Will or of the Intellect

W. Kahl, Die Lehre vom Primat des Willens bei Augustinus, Duns Scotuslind Descartes.

In closest connection with all these general questions stands a spe-

cial psychological problem, which was vigorously discussed through-

out this whole period, and in reference to which the points of

opposition between the parties of the time may be recognised upon a

smaller scale, but all the more sharply focussed. It is the question

whether among the powers of the soul the higher dignity belongs

to the will or to the intellect (utra potentia nobilior). It takes so

broad a space in the literature of this period that the attempt might

have been made to look upon the psychological antithesis which

Unfolds in connection with it as the leading motive of the whole

period. But the course of the development shows too clearly that

the real impelling forces lay in religious metaphysics, and the

rigidity of systematic conception which distinguishes the philoso-

phical doctrines of this period explains sufficiently why it is that

their position with reference to an individual problem may appear

as typical for the different thinkers. It still remains characteristic

that this problem is a question taken from the domain of the inner

world.

1 This doctrine of Occam's concerning secular power and law is followed outto the extreme consequence of the omnipotence of the state by Occam's friend,

Marsilius of Padua, whose treatise, Defensor Pads (1346), carries out in

rigorous lines the attempt to establish the theory of the state upon the utilitarian

tod nominalistic basis using the Epicurean theory of compact (above. §14. f>).

Page 349: Windelband, History of Philosophy, Vol. 1

c'hap. 2, § 26.] Will and Intellect : Thomism, Scotism. 329

In this question, also, the two main bodies of tradition, Augus-

tinianism and Aristotelianism, were not at one ; but their relation was

here in nowise that of an outspoken opposition. For Augustinianism

the question was in general awkwardly stated. For in this system

the oneness of nature in the personality was so strongly emphasised,

and the inter-relation of the different sides of its activity was so

often made prominent, that a relation of rank in the proper sense was

really out of the question. But on the other hand, especially in his

doctrine of knowledge, Augustine had assigned to the will as the

impelling power— even in the process of ideation— a position

so central that it was not shaken in its importance for empirical

facts, even though the Neo-Platonic contemplation of the deity was

maintained as the final goal of development. On the contrary, the

intellectualism of the Aristotelian system was quite undoubted,

and if it still admitted any increase, it had received it from the

Arabian philosophy, especially from Averroism. Thus antitheses

presented themselves which were soon enough to break forth to

open controversy.

Thomism in this point, also, followed Aristotle unconditionally,

finding at its side in this case the nearly related German Mysticism,

and as its opponents the Augustinians, Scotists, and Occamists, so

that, as thus grouped, the opposition between the Dominicans and

the Franciscans finds general expression.

1. The question as to the pre-eminence of the will or of the intel-

lect develops at first as a purely psychological controversy, and de-

mands a decision upon the point, whether in the course of the psychical

life the dependence of the will's decisions upon ideas, or that of the

movements of ideas upon the will, is the greater. It was there-

fore adapted to further the beginnings of a treatment of psychology

that concerned itself especially with the history of mental develop-

ment (cf. § 24), and it would have been able to do this in a higher

degree than was actually the case if it had not always been trans-

ferred to the ground of dialectic or to the metaphysical domain.

This latter transfer occurred principally in consequence of the fact

that the conception of freedom, which always involves ethical and

religious questions, was looked upon as the point in controversy.

Both parties, indeed, desired to maintain or defend man's " freedom "

in the interest of responsibility ; but this was possible only as they

gave different meanings to the word.

Now, in individual cases, Thomas admits an influence of the will,

not only upon motion, but also upon affirmation or denial of ideas.

In particular, he recognises absolutely such an influence in belief.

But in general he regards the will, quite according to the ancient

Page 350: Windelband, History of Philosophy, Vol. 1

330 Mediaeval Philosophy : Second Period. [Part IIL

model, as determined by knowledge of the good. The intellect not

only apprehends in general the idea of the good, but also, in each

individual case, discerns what is good, and thereby determines the

will. The will necessarily strives for that which is known to be

good ; it is therefore dependent upon the intellect. The latter is

the supremus motor of the psychical life; "rationality," so said

Eckhart also, is the head of the soul, and even romantic love

("Ifi'mie") clings only to knowledge. Freedom (as ethical ideal)

is hence, according to Thomas, that necessity which exists upon the

basis of knowledge, and, on the other hand, (psychological) freedom

of choice (facultas electiva) is nevertheless only possible by reason

of the fact that the understanding presents to the will various pos-

sibilities as means toward its end, the will then deciding for that

which is known to be best,— the view held by Albert also. This

intellectualistic determinism, in connection with which Thomas him-

self always insisted that the decision of the will depends only upon

purely internal knowing activities, was extended by his contemporary

Gottfried of Fontaine to the point of making even the sensuous

presentation (phantasma) the causa efficiens of the will's activity.

But the opponents made their attack just in connection with this

conception of necessary determination. The rising of ideas, so

Henry of Ghent had already taught, and after him Duns Scotus, and

still later Occam, is a natural process, and the will becomes un-

avoidably entangled in this if it is to be completely dependent upon

ideas. But with this, said Scotus, contingency (i.e. possibility of

being otherwise or "power to the contrary") in the will's functions

is irreconcilable : for the process of Nature is always determined in

one way ; where it prevails there is no choice. With contingency,

however, responsibility also falls to the ground. Besponsibility can

therefore be preserved only if it is acknowledged that the intellect

exercises no compelling power over the will. To be sure, the co-

operation of the ideational faculty is indispensable in the case of

every activity of the will : it presents the will its objects and the

possibilities of its choice. But it does this only as the servant, and

the decision remains with the master. The idea is never more than

the occasioning cause (causaper accidens) of the individual volition

;

the doctrine of Thomas confuses practical consideration with pure

intellect. If the latter gives the object, the decision is still solely

a matter of the will; the will is the movens per se ; to it belongs

absolute self-determination.

Indeterminism, as Scotus and Occam teach it, sees therefore in the

will the fundamental power of the soul, and maintains conversely,

that as a matter of fact the will on its side determines the develop-

Page 351: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 26.] Will and Intellect : Thomism, Scotism. 331

ment of the intellectual activities. Following the procedure of

Henry of Ghent, 1 according to whom the theoretical functions

become more active according as they are more immaterial, Scotus

attempted to prove the proposition just stated, in a highly interest-

ing manner. The natural process, he says, produces as the first

content of consciousness (cogitatio prima) a multitude of ideas

which are more or less confused (confusaz— indistinctce) and im-

perfect. Of these only those become distinct (distincta) and perfect

on which the will, which in this process is determined by nothing

further, fixes its attention. Scotus also teaches at the same time

that the will strengthens in their intensity these ideas which it

raises from the confused to the distinct condition, and that the

ideas to which the will does not apply itself ultimately cease to

exist, on account of their weakness.

In addition to these psychological arguments, we find appearing

in the controversy appeals to the authority of Anselm and Aristotle

on the one side, and to that of Augustine on the other, and further

a series of other arguments. These are in part of a purely dia-

lectical nature. Such is the case when Thomas claims that

the verum, toward which the intellect aims is higher in rank than the

bonum toward which the will strives, and when Scotus doubts the

authority for this gradation ; and so again when Thomas expresses

the opinion that the intellect apprehends the pure, single conception

of the good, while the will is concerned only with the special

empirical forms assumed by the good, and when Henry of Ghent

and Scotus, exactly reversing this statement, develop the thought

that the will is always directed only toward the good as such, while

the understanding has to show in what the good consists in a

particular case. With such variations the matter was later tossed

to and fro a great deal, and Johannes Buridan is an example of

those who stand undecided between determinism and indeterminism.

For the latter view speaks responsibility, for the lormer the prin-

ciple that every event is necessarily determined by its conditions.

Other arguments which become interwoven in the controversy

trench upon the more general domains of the conceptions of the

world and of life.

2. To this class belongs, first of all, the transfer of the question

of the relative rank of will and intellect to God. The extreme

intellectualism of the Arabians had, in Averroes, excluded the

faculty of will from the Supreme Being, in accordance with the

Aristolelian motif, that every act of will implies a want, a state of

1 Whose view in this respect Richard of Middletown also completely adopted.

Page 352: Windelband, History of Philosophy, Vol. 1

332 Mediaeval Philosophy : Second Period. [Pakt IIL

imperfection and dependence ; on the contrary Avicebron, who ex-

ercised a strong influence upon Duns Scotus, had defended the

religious principle that the world was created by the divine will,

and in a similar line of thought William of Auvergne had main-

tained the originality of the will as existing side by side with the

intellect in the essence of God and in his creative activity. These

antitheses were now continued in the controversy between Thomismand Scotism.

Thomas, indeed, as a matter of course, recognises the reality of

the divine will, but he regards it as the necessary consequence of the

divine intellect, and as determined in its content by the latter. Godcreates only what in his wisdom he knows to be good ; it is neces-

sarily himself, i.e. the ideal content of his intellect, that forms the

object of his will ; he necessarily wills himself, and in this consists

the freedom, determined only by himself, with which he wills indi-

vidual things. Thus the divine will is bound to the divine wisdom,

which is superior to it.

But just in this the opponents of Aquinas see a limitation of

omnipotence which does not comport with the conception of the

ens realissimum. A will seems to them sovereign, only if there is

for it no kind of determination or restriction. God created the

world, according to Scotus, solely from absolute arbitrary will ; he

might have created it, if he had so willed, in other forms, relations,

and conditions ; and beyond this his completely undetermined will,

there are no causes. The will of God with its undetermined crea-

tive resolves is the original fact of all reality, and no further ques-

tions must be asked as to its grounds, — even as the decision made

by the will of a finite being with its liberum arbitrium indifferenticß,

when placed before given possibilities, creates in every instance a

new fact which cannot be understood as necessary.

3. The sharpest formulation of this antithesis comes to light in

the fundamental metaphysical principles of ethics. On both sides

the moral law is naturally regarded as God's command. But

Thomas teaches that God commands the good because it is good,

and is recognised as good by his wisdom ; Scotus maintains that it is

good only because God has willed and commanded it, and Occamadds to this that God might have fixed something else, might have

fixed even the opposite as the content of the moral law. For

Thomas, therefore, goodness is the necessary consequence and mani-

festation of the divine wisdom, and Eckhart also says that "be-

neath the garment of goodness " the essential nature of God is

veiled ; intellectualism teaches the perse'itas boni, the rationalty of

the good. For intellectualism, morals is a philosophical discipline

Page 353: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 2G.] Will and Intellect : Thomism, Scotlsm. 333

whose principles are to be known by the " natural light." " Con-

science " (synteresis J

) is a knowledge of God sub ratione boni. WithScotus and Occam, on the contrary, the good cannot be an object of

natural knowledge, for it might have been otherwise than it is ; it

is determined not by reason, but by groundless will. Nothing, so

Pierre d'Ailly teaches with extreme consistency, is in itself, or

per se, sin; it is only the divine command and prohibition which

make anything such, — a doctrine whose range is understood whenwe reflect that, according to the view of these men, God's com-

mand becomes known to man only through the mouth of the

Church.

It is also closely connected with this that theology, which for

Thomas still remained a "speculative" science, became with his

opponents, as has been already indicated above (§ 25, 3), a "prac-

tical " discipline. Albert had already made intimations of this sort,

Richard of Middletown and Bonaventura had emphasised the fact

that theology deals with the emotions ; Roger Bacon had taught

that while all other sciences are based on reason or experience,

theology alone has for its foundation the authority of the divine

will: Duns Scotus completed and fixed the separation between

theology and philosophy by making it a necessary consequence of

his metaphysics of the will.

4. The same contrast becomes disclosed with like distinctness

in the doctrines of the final destiny of man, of his state in eternal

blessedness. The ancient dcwpia, the contemplation of the divine

majesty, free from will and from want, had in Augustine's teaching

formed the ideal state of the pardoned and glorified man, and this

ideal had been made to waver but little by the doctrines of the ear-

lier Mystics. Now it found new support in the Aristotelian intel-

lectualism, in accordance with which Albert thought that man, in so

far as he is truly man, is intellect. The participation in the divine

being which man attains by knowledge is the highest stage of life

which he can reach. On this account Thomas, too, sets the dianoetic

virtues above the practical, on this account the visio divince essenUce,

the intuitive, eternal vision of God, which is removed beyond all

that is temporal, is for him the goal of all human striving. Fromthis vision follows eo ipso the love of God, just as every determinate

1 This word (written also sindcrrsis, scinderesis) has, since Albert of Boll,

städt, occasioned much etymological cudgelling of brains. Since, however,among the later physicians of antiquity (Sext. Emp.) T^pyjais appears as t>

technical term for "observation," it may be that o-vvr-Zip-qcns, which is attestedin the fourth century, originally signified "self-observation" in analogy withthe Neo-Platonic usage in awaiuO-nins or cwdS^ais (cf. p. 234), and thus tookon the ethico-religious sense of "conscience" (conscientia).

Page 354: Windelband, History of Philosophy, Vol. 1

334 Mediceval Philosophy : Second Period. [Part in-

state of the will is necessarily attached to the corresponding state

of the intellect. Just this tendency of Thomism was given its most

beautiful expression by Dante, the poet of the system. Beatrice is

the poetic embodiment of this ideal, for all time.

Meanwhile a counter-current manifests its force on this point also.

Hugo of St. Victor had characterised the supreme angel choir by

love, and the second by wisdom ; and while Bonaventura regarded

contemplation as the highest stage in the imitation of Christ, he

emphasised expressly the fact that this contemplation is identical

with "love." Duns Scotus, however, taught with a decided polemi-

cal tendency that blessedness is a state of the will, and that, too, of

the will directed toward God alone ; he sees man's last glorification,

not in contemplation, but in love, which is superior to contemplation,

and he appeals to the word of the Apostle, " The greatest of these is

love."

Hence as Thomas regarded the intellect, and Duns Scotus the

will, as the decisive and determining element of man's nature,

Thomas could hold fast to Augustine's doctrine of the gratia irresisti-

bilis, according to which revelation determines irresistibly the intel-

lect and with it the will of man, while Duns Scotus found himself

forced to the "synergistic" view, that the reception of the opera-

tion of divine grace is to a certain extent conditioned by the free will

of the individual. So the great successor of Augustine, with strict

logical consistency, decided against the Augustinian doctrine of pre-

destination.

5. On the other hand, the intellectualism of Thomas develops its

extreme consequences in German Mysticism, whose founder, Eckhart,

is entirely dependent upon the teacher of his Order in the con-

ceptional outlines of his doctrine.1 Eckhart goes far beyond his

master only in the one respect that as a much more original person-

ality he is unwearied in his effort to translate the deep and mighty

feeling of his piety into knowledge, and thus urged on by his inner

nature he breaks through the statutory restrictions before which

Thomas had halted. Convinced that the view of the world given in

the religious cousciousness must be capable of being made also the

content of the highest knowledge, he sublimates his pious faith to a

speculative knowledge, and in contrast with the pure spirituality of

this he looks upon the Church dogma as only the external, temporal

symbol. But while this tendency is one that he shares with many

] Cf. S. Denifle in the Archiv für Litterat.- u Kult.- Gesch. d. M.-A.,II.

417 ff. So far, therefore, as Eckhart was really to be the "Pather of German.sDeculation," this speculation had its source in Thomas Aquinas and his teacher

Albert.

Page 355: Windelband, History of Philosophy, Vol. 1

Cha?. 2, § 26.] Will and Intellect : Eckhart. 334

other systems, it is his peculiarity that he does not wish to have the

inmost and truest truth kept as the privilege of an exclusive circle,

but desires rather to communicate it to all people. He believes

that the right understanding for this deepest essence of religious

doctrine is to be found precisely in connection with simple piety/

and so he throws down from the pulpit among the people the finest

conceptions constructed by science. With a mastery of language

that marks the genius he coins Scholasticism into impressive preach-

ing, and creates for his nation the beginnings of its philosophical

modes of expression,— beginnings which were of determining in*

fluence for the future.

But in his teaching the combined mystical and intellectualistic

elements of Thomism become intensified by the Neo-Platonic ideal-

ism, which had probably reached him through the medium of Scotus

Erigena, to the last logical consequence. Being and knoidedge are

one, and all that takes place in the world is in its deepest essence a

knowing process. The procedure of the world forth out of God is

a process of knowledge, of self-revelation, — the return of things

into God is a process of knowledge, of higher and higher intuition.

The ideal existence of all that is real— so at a later time said

Nicolaus Cusanus, who made this doctrine of Eckhart's his own—is truer than the corporeal existence which appears in space and

time.

The original ground of all things, the deity, must therefore lie

beyond Being and knowledge

;

2it is above reason, above Being ; ii

has no determination or quality, it is " Nothing." But this " deity '•

(of negative theology) reveals itself in the triune God,3 and the

God who is and knows creates out of nothing the creatures whose

Ideas he knows within himself ; for this knowing is his creating.

This process of self-revelation belongs to the essence of the deity:

it is hence a timeless necessity, and no act of will in the propel

sense of the word is required for God to produce the world. Tht

deity, as productive or generative essence, as c' un-natured Nature ;:

[or Nature that has not yet taken on a nature], is real or actual onKby knowing and unfolding itself in God and the world as producec

1 German Mysticism is thus connected with the more general phenomenonthat the fast increasing externalisation which seized upon the life of the Churclin the thirteenth and fourteenth centuries drove piety everywhere into path:

that lay outside the Church.2 Evidently the same relation that subsisted in the system of Plotinus betwee:

the '4v and the voDs, a relation in which thought and Being were held to coincidf.3 The distinction between deity and God (divinitaa and deus) was made di

Jectically by Gilbert de la Porree in connection with the controversy over uuVersals and its relations to the doctrine of the Trinity-

Page 356: Windelband, History of Philosophy, Vol. 1

336 MedicBval Philosophy : Second Period. [Part III.

reality, as natured Nature. 1 God creates all— said Nicolaus Cusa-

nus — that is to say, he is all. Aud on the other hand, according

to Eckhart, all things have essence or substance only in so fai

as they are themselves God ; whatever else appears in them as

phenomena, their determination in space and time, their " here " and

"now " (" Hie " und " Nu" hie et nunc with Thomas), is nothing.2

The human soul, also, is therefore in its inmost nature of the

divine essence, and it is only as a phenomenon in time that it

possesses the variety of " powers " or " faculties " with which it is

active as a member of the natura naturata. That inmost essence

Eckhart calls the " Spark," 3 and in this he recognises the living

point at which the world-process begins its return.

Eor to the "Becoming" corresponds the reverse process, the

"Anti-becoming" ("Entwerten"), the disappearing. And this,

too, is the act of knowledge by means of which the things

which have been made external to the deity are taken back

into the original Ground. By being known by man the world of

sense finds again its true spiritual nature. Hence human cogni-

tion, with its ascent from sense perception to rational insight,4

consists in the "elimination " (" Abscheiden ") of plurality and mul-

tiplicity; the spiritual essence is freed from its enveloping husks.

And this is man's highest task in the temporal life, since knowledge

is the most valuable of man's powers. He should indeed be also

active in this world, and thus bring his rational nature to assert

itself and gain control, but above all outer action, above the right-

eousness of works which belongs to the sphere of sense, stands first

the "inner work," cleanness of disposition, purity of heart, and

above this in turn stands retirement or "decease" (Abgeschieden'

heit) and " poverty " of soul, the complete withdrawal of the soul from

the outer world into its inmost essence, into the deity. In the act

of knowing it reaches that purposelessness of action, that action not

constrained by an end, that freedom within itself, in which its beauty

consists.

But even this is not perfect so long as the knowing process does

not find its consummation. The goal of all life is the knowledge of

1 On the terms natura naturalis and natura naturata, which were probablybrought into use by Averroism (cf. § 27, 1), cf. H. Siebeck, Archiv f. Gesch. d.

Phil., III. 370 ff.

2 Accordingly without accepting the dialectical formulas, Eckhart treats theThomistic doctrine of Ideas quite in the sense of the strict Realism of ScotusErigena. He speaks slightingly of the Nominalists of his time as "little

masters."3 Also the "Gemiithe" or Synteresis = scintilla conscienticc

,

i The single stages of this process are developed by Eckhart according to theThomistic-Augustinian scheme.

Page 357: Windelband, History of Philosophy, Vol. 1

Chap. 2, §27.] Problem of Individuality. 337

God, but knowing is Being; it is a community of life and of Being

with that which is known. Ii' the soul would know God, it must

be God, it must cease to be itself. It must renounce not only sin

and the world, but itself also. It must strip off all its acquired

knowledge, and all present knowing of phenomena; as the deity is

"Nothing," so it is apprehended only in this knowledge that is a

not-knowing— dacht iijnorantla, it was later called by Nicolaus; and

as that " Nothing " is the original ground of all reality, so this not-

knowing is the highest, the most blessed contemplation. It is no

longer an act of the individual, it is the act of God in man ; Godbegets his own essence within the soul, and in his pure eternal

nature the "Spark"' has stripped off all its powers through which it

works in time, and has effaced their distinction. This is the state

of supra-rational knowing when man ends his life in God, — the

state, of which Nicolaus of Cusa said, it is the eternal love (charitas),

which is known by love (amove) and loved by knowledge.

§ 27. The Problem of Individuality.

The doctrine of German Mysticism, which had arisen from the

deepest personal piety and from a genuine individual need felt in

a life whose religion was purely internal, thus runs out into an ideal

of exaltation, of self-denial, of renunciation of the world, in the

presence of which everything that is particular, every individual

reality, appears as sin or imperfection, as had been the case in the

ancient Oriental view. In this thought the contradiction that was

inherent in the depths of the Augustinian system (cf. p. 287) became

fully developed and immediately palpable, and it thus becomes evident

that the Neo-Platonic intellectualism, in whatever form it appeared

from the time of Augustine to that of Master Eckhart, was in itself

alone always necessarily inclined to contest the metaphysical self-

subsistence of the individual, while the other party maintained this

self-subsistence as a postulate of the doctrine of the will. Accord-

ingly, when in connection with the increase of intellectualism the

universalistic tendency increased also, the counter-current was neces-

sarily evoked all the more powerfully, and the same antithesis in

motives of thought which had led to the dialectic of the controversy

over universals (cf. p. 289) now took on a more real and metaphys-

ical form in the question as to the ground of existence in individual

beings (jyrincipiwm individaationis).

1. The stimulus for this was furnished by the far-reaching conse-

quences to which universalism and intellectualism had led amongthe Arabians. For the Arabians, in interpreting the Aristotelian

Page 358: Windelband, History of Philosophy, Vol. 1

338 Mediaeval Philosophy : Second Period. [Pakt 111

system, had proceeded in the direction which had been introduce*

in antiquity by Strato (cf. p. 179 f.), and which among the later commentators had been maintained chiefly by Alexander of Aphrodisias

This direction was that of naturalism, which would fain remove

from the system of the Stagirite even the last traces of a metaphys-

ical separation between the ideal and the sensuous. This effort had

become concentrated upon two points : upon the relation of God to.

the world, and upon that of the reason to the other faculties. In

both these lines the peculiar nature of the Arabian Peripatetic doc-

trine developed, and this took place by complicated transformations

of the Aristotelian conceptions of Form and Matter.

In general, we find in this connection in the Andalusian philoso-

phy a tendency to make matter metaphysically self-subsistent. It

is conceived of, not as that which is merely abstractly possible, but

as that which bears within itself as living germs the Forms peculiar

to it, and brings them to realisation in its movement. At the same

time Averroes, as regards particular cosmic processes, held fast to

the Aristotelian principle that every movement of matter by which

it realises out of itself a lower Form, must be called forth by a

higher Form, and the graded series of Forms finds its termination

above in God, as the highest and first mover. The transcendence

of God could be united with this view, as the doctrine of Avicebron

shows, only if matter were regarded as itself created by the divine

will. But on the other hand, this same Jewish philosopher, pro-

ceeding from the same presuppositions, insisted that with the excep-

tion of the deity, no being could be thought of otherwise than as

connected with matter, that accordingly even the spiritual Forms

need for their reality a matter in which they inhere, and that finally

the living community of the universe demands a single matter as

basis for the entire realm of Forms. The more, however, in the

system of Averroes, matter was regarded as eternally in motion

within itself, and as actuated by unity of life, the less could the

moving Form be separated from it realiter, and thus the same divine

All-being appeared on the one hand as Form and moving force

{natura naturans) , and on the other hand as matter, as moved world

(natura naturata).

This doctrine with regard to matter, that it is one in nature, i&

informed within, and is eternally in motion of itself became ex-

tended with Averroism as an extremely naturalistic interpretatiou

of the philosophy of Aristotle. It now became reinforced by those

consequences of dialectical Eealism which compelled the view thai

God, as the ens generalissimum, is the only substance, and that in-

dividual things are but the more or less transient Forms in whicl

Page 359: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 27.] Problem of Individuality : Averroes. 339

t'ais single substance becomes realised (cf. § 23). The Amalricans

Mius teach that God is the one single essence (essentia) of all things,

and that creation is only an assuming of form on the part of this

divine essence, a realising, completed in eternal movement, of all

possibilities contained in this one single matter. David of Dinant 1

establishes this same pantheism with the help of Avicebron's con-

ceptions, by teaching that as "hyle" (i.e. corporeal matter) is the

substance of all bodies, so mind (ratio—- mens) is the substance of

all souls ; that, however, since God, as the most universal of all es-

sences, is the substance of all things whatever, God, matter, and

mind are, in the last resort, identical, and the world is but their

self-realisation in particular forms.

2. But the metaphysical self-subsistence of the individual mindwas involved in doubt by yet another line of thought. Aristotle

had made the vous, as the everywhere identical rational activity,

join the animal soul " from without," and had escaped the difficul-

ties of this doctrine because the problem of personality, which

emerged only with the Stoic conception of the ^ye/xoviKov, did not

as yet lie within the horizon of his thought. But the commenta-

tors, Greek and Arabian, who developed his system did not shrink

before the consequences that resulted from it for the metaphysical

value of mental and spiritual individuality.

In the thought of Alexander of Aphrodisias we meet, under the

name of the "passive intellect" (cf. p. 150), the capacity of the in-

dividual psyche to take up into itself, in accordance with its whole

animal and empirical disposition, the operation of the active reason,

and this intellectus agens (agreeably to the naturalistic conception of

the whole system) is here identified with the divine mind, which is

still thought only as "separate Form" (intellectus separatus). But

with Simplicius, in accordance with the Neo-Platonic metaphysics, this

intellectus agens which realises itself in man's rational knowledge

has already become the lowest of the intelligences who rule the sub-

lunary world. 2 This doctrine finds an original development in the

thought of Averroes.3 According to his view, the intellectus passivus

is to be sought in the individual's capacity for knowledge, a capacity

which, like the individual himself, arises and perishes as Form of

the individual body ; it has validity, therefore, only for the indi-

vidual, ar.d for that which concerns the particular. The intellectus

1 Following the Liber de Causis and the pseudo-Boethian treatise i> TJno et

Unitcae ; cf.B. Haureau in the Mimoires de V Acad, des Inscript., XXIX. (l*s 77),

and also A. Jundt, Histoire du PantMisme Populaire au M.-A. (Paris, 1876).2 The so-called "Theology of Aristotle" identifies this i-ofs with the \07cn.

For particulars, see E. Renan, Av. et /Mr., [I. § ß ff.

8 Cf. principally his treatise Be Animce Beatitudine.

Page 360: Windelband, History of Philosophy, Vol. 1

340 Mediaeval Philosophy : Second Period. [Part III.

agens, on the contrary, as a Form existing apart from empirical in-

dividuals and independent of them, is the eternal generic reason of

the human race, which neither arises nor perishes, and which con-

tains the universal truths in a manner valid for all. It is the sub-

stance of the truly intellectual life, and the knowing activity of

the individual is but a special manifestation of it. This (actual)

knowing activity (as intellectus acquisitus) is indeed in its con-

tent, in its essence, eternal, since in so far it is just the active rea-

son itself ; on the contrary, as empirical function of an individual

knowing process, it is as transitory as the individual soul itself.

The completest incarnation of the active reason has, according to

Averroes, been given in Aristotle. 1 Man's rational knowing is,

then, an impersonal or supra-personal function : it is the individual's

temporal participation in the eternal generic reason. This latter is

the unitary essence which realises itself in the most valuable activi-

ties of personality.

Intimations of this pan-psychism occasionally appear in the train

of Neo-Platonic Mysticism at an earlier period in Western literature

;

as an outspoken and extended doctrine it appears by the side of

Averroism about 1200 ; the two are everywhere named in conjunc-

tion at the first when the erroneous doctrines of the Arabian

Peripatetic thought are condemned, and it is one main effort of

the Dominicans to protect Aristotle himself from being confused

with this doctrine. Albert and Thomas both write a De Unitate

Intellectus against the Averroists.

3. Pan-psychism encounters with Christian thinkers an oppo-

sition in which the determining factor is the feeling of the meta-

physical value of personality,— the feeling which had been nour-

ished by Augustine. This is the standpoint from which men like

William of Auvergne and Henry of Ghent oppose Averroes. Andthis is also the real reason why the main systems of Scholasticism

— in diametrical contrast with Eckhart's Mysticism— did not allow

the Realism which was inherent in the intellectualistio bases of

their metaphysics to come to complete development. Thomismwas here in the more difficult case, for it maintained indeed, follow-

ing Avicenna's formula (cf. p. 299), that universals, and therefore

also the genus " soul," exist only " individualised," i.e. in the indi-

vidual empirical examples as their universal essence (quidditas),

but it ascribed to them, nevertheless, metaphysical priority in the

divine mind. It was therefore obliged to explain how it comes

: And with this the unconditional recognition of the authority of the Stagirite

is theoretically justified by Averroes.

Page 361: Windelband, History of Philosophy, Vol. 1

Chap. 2, § -27.] Problem of Individuality : Thomism, Scotism. 841

about that this one essence as universal matter presents itself in

such manifold forms. That is to say, it asked after the princijpium

individuations, and found it in the consideration that matter in

space and time is quantitatively determined {materia signata). In

the capacity of matter to assume quantitative differences consists the

possibility of individuation, i.e. the possibility that the same Form{e.g. humanity) is actual in different instances or examples as indi-

vidual substances. Hence, according to Thomas, pure Forms {sepa-

ratee, sive subsistentes) are individualised only through themselves

;

that is, there is but one example which corresponds to them. Every

angel is a genus and an individual at the same time. The inherent

Forms, on the contrary, to which the human soul also belongs in

spite of its subsistence (cf. p. 324). are actual in many examples, in

accordance with the quantitative differences of space and time

which their matter presents.

This view was opposed by the Franciscans, whose religious and

metaphysical psychology had developed in intimate relation with

Augustine's teaching. In their thought, first the individual soul,

and then, with a consistent extension in general metaphysics,

individual beings in general, are regarded as self-subsisting realities.

They rejected the distinction of separate and inherent Forms.

Bonaventura, Henry of Ghent, and still more energetically Duns

Scotus, maintained, following Avicebron, that even intellectual

Forms have their own matter, and Scotus teaches that the " soul " is

not individualised and substantialised only after, and by means of,

its relation to a definite body, as Thomas had taught, but that it is

already in itself individualised and substantialised. On this point

Scotism shows a discord which had evidently not come to notice in

the mind of its author. It emphasises on the one hand, in the

strongest manner, the Reality of the universal, by maintaining the

unity of matter {materia primo-prima) quite in the Arabian sense,

and on the other hand it teaches that this universal is only actual

by being realised by the series of Forms descending from the uni-

versal to the particular, and ultimately by means of the definite

individual Form {hcecceitas) . This individual Form is therefore

for Duns Scotus an original fact; no farther question as to its

ground is permissible. He designates individuality (both in the

sense of individual substance and in that of individual occurrence)

as the contingent {contingens) ; that is, as that which is not to be

deduced from a universal ground, but is only to be verified as actual

fact. For him, therefore, as for his predecessor Roger Bacon, the

inquiry for the principle of individuation has no meaning : the indi-

vidual is the "last" Form of all reality, by means of which alone

Page 362: Windelband, History of Philosophy, Vol. 1

842 Mediceval Philosophy : Second Period. [Part III

universal matter exists, and the question rather is. how, in presence

of the fact that the individual being with its determined form is

the only Reality, one can still speak of a Reality of universal

" natures." 1

From this noteworthy limitation of the doctrine of Scotus it

becomes explicable that while some of its adherents, as for example

Francis of Mayron, proceeded from it to extreme Realism, it sud-

denly changed with Occam into the renewal of the nominalistic

thesis, that only the individual is real and that the universal is but

a product of comparative thought.

4. The victorious development which Nominalism experienced in

the second period of medieeval philosophy rests upon an extremely

peculiar combination of very different motives of thought. In the

depths of this stream of development is dominant the Augustinian

moment of feeling, which seeks to see the proper metaphysical value

secured to the individual personality; in the main philosophical

current the anti-Platonic tendency of the Aristotelian theory of

knowledge, now just becoming known, asserts itself, throwing its

influence toward conceding the value of "first substance" to the

empirical individual only; and on the surface plays a logico-gram-

matical schematism, which has its origin in the first operation of

the Byzantine tradition of ancient thought.2 All these influences

become concentrated in the impassioned, impressive personality of

William of Occam.

In their exposition of the doctrine of concepts and its application

to the judgment and syllogism, the text-books of " modern " logic,

as type of which that of Petrus Hispanus may serve, lay an impor-

tant emphasis upon the theory of " supposition " in a manner which

is not without its precedent in antiquity.3 According to this theory

a class-concept or term (terminus) may, in language, and, as was

then supposed, in logic also, stand for the sum of its species, and a

species-concept for the sum of all its individual examples (homo =omnes homines), so that in the operations of thought a term is

employed as a sign for that which it means. Occam develops Nom-inalism in the forms of this Terminism* (cf. pp. 325 f). Individual

1 This method for the solution of the problem of universals, peculiar to DunsScotus, is usually called Formalism.

2 In fact, we may see in the working of the text-book of Michael Psellos the

first impetus of that accession of ancient material of culture which the Westreceived by way of Byzantium, and which later in the Renaissance becamedefinitely united with the two other lines of tradition that came, the one byway of Rome and York, the other by way of Bagdad and Cordova.

3 The reader need only be reminded of the investigations of Philodemus onsigns and things signified (p. 162 ; cf. also p. 198).

* Cf. K. Prantl in the Süz.-Ber. der Münch. Acad.. 1864, II. a 58 ff.

Page 363: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 27.] Problem of Individuality : Terminism. 343

things, to which Occam, following Scotus, concedes the lleality of

original Forms, are represented in thought by us intuitively, without

the mediation of species intelligibiles ; but these ideas or mental rep-

resentations are only the " natural " signs for the things represented.

They have only a necessary reference to them, and have real simi-

larity with them as little as any sign is necessarily like the object

designated. This relation is that of " first intention." But now as

individual ideas stand for (supponunt) individual things, so, in

thought, speech, and writing, the " undetermined " general ideas of

abstract knowledge, or the spoken or written words which in turn

express these general ideas, may stand for the individual idea. This" second intention," in which the general idea with the help of the

word refers no longer directly to the thing itself, but primarily to the

idea of the thing, is no longer natural, but arbitrary or according

to one's liking (adplacitum instituta). 1 Upon this distinction Occamrests also that of real and rational science : the former relates imme-diately or intuitively to things, the latter relates abstractly to the

immanent relations between ideas.

It is clear, according to this, that rational science also presupposes

" real " science and is bound to the empirical material presented in

the form of ideas by this real science, but it is also clear that even" real " knowledge apprehends only an inner world of ideas, which

may indeed serve as " signs " of things, but are different from things

themselves. The mind— so Albert had incidentally said, and Nico-

laus Cusanus at a later time carried out the thought— knows only

what it has within itself; its knowledge of the world, terministic

Nominalism reasons, refers to the inner states into which its living

connection with the real world puts it. As contrasted with the true

essence of things, teaches Nicolaus Cusanus, who committed himself

absolutely to this idealistic Nominalism, human thought possesses

only conjectures, that is, only modes of representation which corre-

spond to its own nature, and the knowledge of this relativity of all

positive predicates, the knowledge of this non-knowledge, the docta

ignorantia, is the only way to go beyond rational science and attain

to the inexpressible, signless, immediate community of knowledge

with true Being, the deity.

5. In spite of this far-reaching epistemological restriction, the

real vital energy of Nominalism was directed toward the develop-

ment of natural science ; and if its results during the fourteenth and

fifteenth centuries remained very limited, the essential reason for this

1 The agreement of this with the contrast between 0/cm and <pv<ris, which hadbeen asserted also in the ancient philosophy of language (Plato's Cratylus),is obvious.

Page 364: Windelband, History of Philosophy, Vol. 1

344 Mediaeval Philosophy : Second Period. [Part III

was that the scholastic method with its bookish discussion of authori-

ties, which had now attained full perfection, controlled absolutely

later as well as earlier the prosecution of science, and that the

new ideas forced into this form could not unfold freely,— a phe-

nomenon, moreover, which continues far into the philosophy of the

Renaissance. For all that, Duns Scotus and Occam gave the chief

impetus to the movement in which philosophy, taking its place

beside the metaphysics whose interests had hitherto been essentially

religious, made itself again a secular science of concrete, actual fact,

and placed itself with more and more definite consciousness upon

the basis of empiricism. When Duns Scotus designated the hcecceitas

or original individual Form, as contingent, this meant that it was to

be known, not by logical deduction, but only by actual verification

as fact ; and when Occam declared the individual being to be the

alone truly Real, he was thereby pointing out to " real science " the

way to the immediate apprehension of the actual world. But in

this point the two Franciscans are under the influence of Roger

Bacon, who with all his energy had called the science of his time

from authorities to things, from opinions to sources, from dialectic

to experience, from books to Nature. At his side in this movementstood Albert, who supported the same line of thought among the

Dominicans, knew how to value the worth of original observation

and experiment, and gave brilliant proof in his botanical studies

of the independence of his own research. But strongly as Roger

Bacon, following Arabian models, urged quantitative determinations

in observation, and mathematical training, the time was not yet

ripe for natural research. Attempts like those of Alexander

Xekkam (about 1200), or those of Nicolaus d'Autricuria, at a later

time (about 1350), passed away without effect.

The fruitful development of empiricism during this period was

only in the line of psychology. Under the influence of the Arabs,

especially of Avicenna and of the physiological optics of Alhacen,

investigations concerning the psychical life took on a tendency

directed more toward establishing and arranging the facts of expe-

rience. This had been begun even by Alexander of Hales, by his

pupil, Johann of Rochelle, by Vincent of Beauvais, and especially

by Albert; and in the system of Alfred the Englishman (Alfred de

Sereshel, in the first half of the thirteenth century) we find a

purely physiological psychology with all its radical consequences.

These stirrings of a physiological empiricism would, however, have

been repressed by the metaphysical psychology of Thomism, if they

had not found their support in the Augustinian influence, which

held fast to the experience which personality has of itself, as its

Page 365: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 27.] Problem of Individuality : Nicolaus Gusanus. 345

highest principle. In this attitude Henry of Ghent, especially, cameforward in opposition to Thoinism. He formulated sharply the

standpoint of inner experience and gave it decisive value, particu-

larly in the investigation of the states of feeling. Just in this

point, in the empirical apprehension of the life of feeling, the

theory of which became thus emancipated at the same time from

that of the will and that of the intellect, he met support in Roger

Bacon, who, with clear insight and without the admixture of meta-

physical points of view, distinctly apprehended the difference in

principle between outer and inner experience.

Thus the remarkable result ensued, that purely theoretical science

developed in opposition to intellectualistic Thomism, and in connec-

tion with the Augustinian doctrine of the self-certainty of person-

ality. This self-knowledge was regarded as the most certain fact of

"real science," even as it appeared among the nominalistic Mystics

such as Pierre d'Ailly. Hence " real science " in the departing

Middle Ages allied itself rather to active human life than to Nature;

and the beginnings of a " secular " science of the inter-relations of

human society are found not only in the theories of Occam and

Marsilius of Padua (cf. p. 328), not only in the rise of a richer,

more living, and more " inward " writing of history, but also in an

empirical consideration of the social relations, in which a Nicolas

cV Oresme, 1 who died 1382, broke the path.

6. The divided frame of mind in which the departing Middle

Ages found itself, between the original presuppositions of its

thought and these beginnings of a new, experientially vigorous

research, finds nowhere a more lively expression than in the phil-

osophy of Nicolaus Gusanus, which is capable of so many interpre-

tations. Seized in every fibre of his being by the fresh impulse of

the time, he nevertheless could not give up the purpose of arrang-

ing his new thoughts in the system of the old conception of the

world.

This attempt acquires a heightened interest from the conceptions

which furnished the forms in which he undertook to arrange his

thoughts. The leading motive is to show that the individual, even

in his metaphysical separateness, is identical with the most uni-

versal, the divine essence. To this end Nicolaus employs for the

first time, in a thoroughly systematic way, the related conceptions

of the infinite and the finite. All antiquity had held the perfect to

be that which is limited within itself and had regarded only

indefinite possibility as infinite. In the Alexandrian philosophy,

Cf. concerning him W. Roscher, Zeitschr. f. Staatswissenschafi, 1803, 305 ff.

Page 366: Windelband, History of Philosophy, Vol. 1

346 Mediceval Philosophy : Second Period. [Fakt HJ

on the contrary, the highest being was stripped of all finite at

tributes. In Plotinus the " One " as the all-forming power is

provided with an unlimited intensity of Being on account of the

infinity of matter in which it discloses itself ; and also in Christian

thought the power, as well as the will and the knowledge of God,

had been thought more and more as boundless. Here the main

additional motive was, that the will even in the individual is felt

as a restless, never quiet striving, and that this infinity of inner ex-

perience was exalted to a metaphysical principle. But jSficolaus was

the first to give the method of negative theology its positive ex-

pression by treating infinity as the essential characteristic of God in

antithesis to the world. The identity of God with the world,

required as well by the mystical view of the world as by the

naturalistic, received, therefore, the formulation that in God the

same absolute Being is contained infinitely, which in the world

presents itself in finite forms.

In this was given the farther antithesis of unity and plurality.

The infinite is the living and eternal unity of that which in the

finite appears as extended plurality. But this plurality— and

Cusanus lays special weight on this point— is also that of opposites.

What in the finite world appears divided into different elements,

and only by this means possible as one thing by the side of anothei

in space, must become adjusted and harmonised in the infinitude

of the divine nature. God is the unity of all opposites, the coin-

cidentia oppositorum. 1 He is, therefore, the absolute reality in

which all possibilities are eo ipso realised Rossest, can-is), while

each of the many finite entities is in itself only possible, and is real

or actual only through him. 2

Among the oppositions which are united in God, those between him

and the world, — that is, those of the infinite and the finite, and oi

unity and plurality,— appear as the most important. In consequence

of this union the infinite is at the same time finite ; in each of his

manifestations in phenomena the unitary deus implicitus is at the

same time the deus explicitus poured forth into plurality (cf. p.

290). God is the greatest (maximum) and at the same time also

1 Nicolaus also designates his own doctrine, in contrast with opposing sys-

tems, as a coincidentia oppositorum, since it aims to do justice to all motives of

earlier philosophy. Cf. the passages in Falckenberg, op. cit., pp. 60 ff.

2 Thomas expressed the same thought as follows : God is the only necessarybeing, i.e. that which exists by virtue of its own nature (a thought which is to beregarded as an embodiment of Anselm's ontological argument, cf

; § 23, 2), whilein the case of all creatures, essence (or quidditas— whatness) is reahy separatefrom existence in such a way that the former is in itself merely possible andthat the latter is added to it as realisation. The relation of this doctrine to thefundamental Aristotelian conceptions, actus and potentia, is obvious.

Page 367: Windelband, History of Philosophy, Vol. 1

Chap. 2, § 27.] Problem of Individuality : Nicolaus öusanus. 34?

the smallest (minimum). But, on the other hand, in consequence

of this union it follows also that this smallest and finite is in its

own manner participant in the infinite, and presents within itself,

as does the whole, a harmonious unity of the many.

Accordingly, the universe is also infinite, not indeed in the same

sense in which God is infinite, but in its own way; that is, it is

unlimited in space and time (interminatum, or privitively infinite).

But a certain infinity belongs likewise to each individual thing,

in the sense that in the characteristics of its essence it carries

within itself also the characteristics of all other individuals. All

is in all: omnia ubique. In this way every individual contains

within itself the universe, though in a limited form peculiar to this

individual alone and differing from all others. In omnibus partibus

relucet totum. Every individual thing is, if rightly and fully known,

a mirror of the universe, — a thought which had already been ex-

pressed incidentally by the Arabian philosopher Alkendi.

Naturally this is particularly true in the case of man, and in his

conception of man as a microcosm Nicolaus attaches himself

ingeniously to the terra inistic doctrine. The particular manner in

which other things are contained in man is characterised by the

ideas which form in him signs for the outer world. Man mirrors

the universe by his "conjectures," by the mode of mental repre-

sentation peculiar to him (cf. above, p. 343).

Thus the finite also is given with and in the infinite, the individ-

ual with and in the universal. At the same time the infinite is

necessary in itself; the finite, however (following Duns Scotus), is

absolutely contingent, i.e. mere fact. There is no proportion

between the infinite and the finite ; even the endless series of the

finite remains incommensurable with the truly infinite. The deri-

vation of the world from God is incomprehensible, and from the

knowledge of the finite no path leads to the infinite. That which

is real as an individual is empirically known, its relations and the

oppositions prevailing in it are apprehended and distinguished by

the understanding, but the perception or intuition of the infinite

unity, which, exalted above all these opposites, includes them all

within itself, is possible only by stripping off all such finite knowl-

edge, by the mystical exaltation of the docta ignorantia. Thus the

elements which Cusanus desired to unite fall apart again, even in

the very process of union. The attempt to complete the mediaeval

philosophy and make it perfect on all sides leads to its inner

lisintegration.

Page 368: Windelband, History of Philosophy, Vol. 1
Page 369: Windelband, History of Philosophy, Vol. 1

APPENDIX.

P. 12. Line 15. Add :—

On the pragmatic factor, cf. C. Herrmann, Der pragmatische Zusammenhangin der Geschichte der Philosophie (Dresden, 1863).

P. 12. Line 10 from foot of the text. Add as foot-note, affixed

to the word " positive ":—

A similar, but quite mistaken attempt lias been recently made in this direc-

tion by Fr. Brentano, Die vier Phasen in der Philosophie und ihr gegenwärtigerStand (Vienna, 1895). Here belong also the analogies, always more or less

artificial, which have been attempted between the course of development in theancient and that in the modern philosophy. Cf. e.g. v. Reichlin-Meldegg, DerParallelismus der alten und neueren Philosophie (Leips. and Heidelb. 1865).

P. 16. Line 6 from foot of text, add :—

In all previous expositions of the history of philosophy, whether upon a largeror smaller scale, a chronological arrangement has been adopted, following theorder and succession of the more important philosophies and schools. Thesevarious arrangements have differed only in details, and these not always impor-tant. Among the most recent might be named in addition, that of J. Bergmann,whose treatment shows taste and insight (2 vols., Berlin, 1892). A treatmentmarked by originality and fineness of thought, in which the usual scheme hasbeen happily broken through by emphasis upon the great movements and inter-

relations of the world's history, is presented by K. Eucken, Die Lebensanschau*ungen der grossen Denker (2d ed., Leips. 1898).

P. 23. To the foot-note, add :—

Windischmann, earlier (Die Philosophie im Fortgang der Weltgeschichte,Bonn, 1827-1834), and recently P. Deussen (Allgemeine Geschichte der Philoso-

phie, I. 1, Leips. 1894) have made a beginning toward the work of relating this

Oriental thought to the whole history of philosophy.

P. 24. Line 8. Affix as foot-note :—

E. Rohde has set forth with great insight and discrimination the rich sugges-

tions for philosophy in the following period, which grew out of the transforma-tions of the religious ideas (Psyche, 2d ed., 1897).

P. 27. To the lit. on the Period, add :—

A. Fairbanks, The First Philosophers of Greece, N.Y. 1898.

P. 30. Line 30. To the notice of Heraclitus, add :—

He was apparently the first who, from the standpoint of scientific insight,

undertook to reform the public life and combat the dangers of anarchy. Him-self an austere and rigorous personality, he preached the law of order, which

ought to prevail in human life as in nature.

349

Page 370: Windelband, History of Philosophy, Vol. 1

350 Appendix.

P. 30. Line 19 from the foot. To the notice of Anaxagoras, add :—His scientific employments were essentially astronomical in their nature.

Neglecting earthly interests, he is said to have declared the heavens to be his

fatherland, and the observation of the stars to be his life work. Metrodorus andArchelaus are named as his disciples.

P. 42. Foot-note 1. Kelating to the wvs of Anaxagoras, add :—Cf., however, M. Heinze in the Ber. d. Sachs. Ges. d. Wiss., 1890.

P. 46. Last line of text. To the word " curved," affix as foot-

note :—

The tradition (Arist., loc. cit.) shows this collocation ; whereas, from thecosmology of the Pythagoreans and likewise from that of Plato and Aristotle, weshould expect the reverse order.

P. 55. To the notice of Diogenes of Apollonia, add :—

He was the most important of the eclectics of the fifth century. So little is

known as to his life that it is even doubtful whether Apollonia was his home.Of his writings, even Simplicius had only the irepi <p6crews before him (Phys.,32 V. 151, 24 D).

P. 62. Add to foot-note 1 :—

because in this phase of Greek thought they run along as yet unrelated lines ofthought, side by side with the theories of natural science. Only the Pythago-reans seem as yet to have begun the combination between theology and phi-losophy, which later became through Plato a controlling influence.

P. 68. Prefix to par. 4, which begins with "But while," the

following sentence :—

A preparation for this transition was made by the circumstance

that even in the investigation of nature, interest in fundamental

principles had grown weaker after the first creative development,

and science had begun to scatter her labours over special fields.

P. 71. To the personal notice of Socrates, add :—

He considered this enlightenment of himself and fellow-citizens a divine voca-tion (Plato's Apology), giving this work precedence even over his care of his

family (Xanthippe). He gathered about him the noblest youth of Athens, suchas Alcibiades, who honoured in him the ideal and the teacher of virtue. Heappeared thus as leader of an intellectual aristocracy, and just by this meanscame into opposition to the dominant democracy. [K. Joel, Der echte u. d.

Xenophontische Sokrates, Vol. I., Berlin, 1893. Vol. II. in 2 pts., 1901. Kralik,

Sokrates, 1899.]

P. 96. Line 23. Insert after Plato :—

And of their materialism which he so vigorously opposed.

P. 102. At close of par. 4, insert :—

This personal influence he himself regarded as the most important part of his

activity. For scientific investigation was only one side of his rich nature. Thedemand for ethical teaching and for political and social efficiency had a still

stronger life within him. He had an open vision for the evils of his time. Heunited an adherence to the aristocratic party with an activity in the directioa

indicated by Socrates, and never quite gave up the hope of reforming the life of

his time through his science. To this was added as a third element in his per-

sonality that pre-eminent artistic disposition which could clothe his ideals withpoetic exposition in the most splendid language.

Page 371: Windelband, History of Philosophy, Vol. 1

Appendix. 351

P. 103. To references on Plato, add :—

P. Lutowslawski, Origin and Growth of Plato's Logic (1897).

[R. L. Nettleship, Philos. Lectures, ed. by Bradley and Benson, 1897. W.Windelband, Plato, Stuttgart, 1900.]

P. 104. After first par., insert :—

In comparison with the high flight of Plato, the personality and life-work of

Aristotle appear throughout of cooler and soberer type. But if he lacks the

impulse toward an active influence in public life, and also the poetic charm of

diction and composition, he has, instead, all the more effective a substitute in

the power of thought with which he surveys and masters his field, in the clarity

and purity of his scientific temper, in the certainty and power with which hedisposes and moulds the results gathered from the intellectual labours of manycontributors. Aristotle is an incarnation of the spirit of science such as the

world has never seen again, and in this direction his incomparable influence haslain. He will always remain the leading thinker in the realm of investigation

which seeks to comprehend reality with keen look, unbiassed by any interest

derived from feeling.

P. 104. Line 10. After " knowledge," insert :—

The recently discovered main fragment of his UoXirela twv 'Advvaluv is a valu-

able example of the completeness of this part, also, of his literary work. In themain only his scientific, etc.

P. 104. [Especially valuable in the recent literature upon Aristotle are : H.Meier, Die Syllogistik des Aristoteles. Vol. I., 1896, Vol. II. in 2 pts., 1900 ; G.

Rodier, Aristote, Traite de VAme, trad, et annotee. 2 vols., Paris, 1900. Cf. also

W. A. Hammond, ASs Psychology : The De Anima and Parva Nat., tr. withInt. and Notes, Lond. and N.Y. 1901 ; H. Siebeck, A., Stuttgart, 1899.]

P. 112. As note to close of first par., attached to words " in the

middle ":—

Cf., however, on this, A. Goedeke-Meyer, Die Naturphilosophie Epikur's inihrem Verhältniss zu Demokrit, Strassburg, 1897.

P. 119. Line 17. After " back," insert :—

according to the general laws of association and reproduction

(Phaedo, 72 ff.).

P. 123. Insert after the first par. under 6, the following par. :—

This completely new attempt on Plato's part was supported by the

theological doctrines which he was able to take from the Mysteries of

Dionysus. Here the individual soul was regarded as a " daimon " or

spirit which had journeyed or been banished from another world into

the body, and during its earthly life maintained mysterious emo-

tional relations to its original home. Such theological ideas were

brought by the philosopher into his scientific system, not without

serious difficulties.

P. 135. Kote attached to the word "not" in line 11 (from

foot) :—

For Aristotle means nothing else, even where, as is frequently the case in theAnalytics, he expresses the relation by saying that the question is whether the

one concept is affirmed or predicated (Karwyopeiv) of the other.

Page 372: Windelband, History of Philosophy, Vol. 1

352 Appendix.

P. 142. After the first sentence in the last par., insert :—

" The subordination of the single thing under the general concept

is for him too, not an arbitrary act of the intellect in its work of

comparison; it is an act of knowledge which takes us into the

nature of things and reproduces the actual relations which obtain

there."

P. 148. Line 3. After " world," insert :—

Every element has thus its " natural " motion in a certain direc-

tion and its " natural " place in the universe. Only by collision with

others (/Jia) is it turned aside or crowded out.

P. 162. Before second par., insert :—

" In the history of the Stoa we have to distinguish an older period which waspredominantly ethical, a middle period which was eclectic, and a later periodwhich was religious."

P. 162. To references on Stoicism, add :—

A. Schmekel, Die mittlere Stoa (Berlin, 1892).

P. 162. Line 6 from foot. To references on Lucretius, add :—

R. Heinze's Com. on 3d Book (Leips. 1877).

P. 163. Line 20. Add :—

Cf. E. Pappenheim (Berlin, 1874 f., Leips. 1877 and 1881).

P. 163. To references on Scepticism, add :—

V. Brochard, Les Sceptiques Ghrecs (Paris, 1887). [M. M. Patrick, SextusEmpiricus and Greek Scepticism (contains trans, of the " Pyrrhonic Sketches,"Camb. and Lond. 1899).]

P. 163. Line 35. After " principle," insert :—

Cicero stands nearest to the position of Probabilism as maintained by the

Academy. See below, § 17, 7.

P. 163. To the material before § 14, add :—

A popular moral eclecticism was represented by certain preachers of moralswho were more or less closely related to the principles of the Cynics. Thesescourged the social and moral conditions of the Hellenistic and later of the

Roman world with harsh and outspoken criticism. Among them were Teles (cf.

v. Wilamowitz-Möllendorf, Philologische Untersuchungen, IV., 292 ff. ; Frag-ments, ed. by 0. Hense, Freiburg, 1899), Bion of Borysthenes (cf. R. Heinze,

de Horatio Bionis Imitatore, Bonn, 1889) of a later period, Demetrius, Oeno-maos, and Demonax. Cf. J. Bernays, Lukian und die Kyniker (Berlin, 1879).

In this connection Dio Chrysostomos is also to be named. Cf. H. v. Arnim(Berlin, 1898).

P 174. Line 8. Add to this paragraph :—

In many cases, however, notably in the Imperial age of Rome,

this maxim appears as the easily intelligible principle of the honour-

able man who finds himself repelled by the corruption and partisan

seif-seeking of political life, and can have nothing to do with it.

Page 373: Windelband, History of Philosophy, Vol. 1

Appendix. 353

P. 181. Add to the second par. the following (in part new) : —Nevertheless, inasmuch as they, like Heraclitus, treated the neces-

sary course of events and providence as equivalent terma, the Stoic

formulation of the principle of sufficient reason (i.e. that everything

which comes to be has a ground or reason) may also be expressed in

the form that not even the least thing in the world can be otherwise

than in accord with the decree of Zeus.

P. 186. Line 8 from foot of text, after "Heraclitus" insert:—"and in part to the later philosophy of nature as influenced byhim. (Pseudo-Hippoc. vrept oWt^s ; cf. above p. 67, note 1.)

P. 189. Line 12 from foot, add the following :—

Finally this web of syncretistic theology received the metaphysi-

cal strand, to which the Older Academy with Pythagorean tenden-

cies (especially Xenocrates) had begun to attach the hierarchy of

mythical forms (cf. § 11, 5). The combination of all these theo-

logical tendencies was completed in the middle, eclectic Stoa, espe-

cially through Posidonius.

P. 204. Note 4, add :—

Hence Epicurus did not regard it necessary to decide on theoretical groundsbetween different modes of explaining particular phenomena : the one mode wasno more valid (ov päWov) than the other, to use the sceptical phrase.

P. 210. Line 20. Add :—

trans, as Harnack's History of Doctrine, by N. Buchanan, Lond. 1894.

P. 210. Add to references :—

Fr. Susemihl, Geschichte der griechischen Litteratur in der Alexandrinerzeit(2 vols., Leips. 1891).

P. 216. Line 26. To the lit., add :—

H. v. Arnim, Dion von Prusa (Leips. 1896), pp. 4-114.

P. 216. Line 16 from foot. To the notice of Galen, add :—

He was frequently referred to as philosophical authority in the humanisticliterature of the Renaissance. His treatise, De placitis Hippocratis et Piatonis,has been edited by J. Müller (Leips. 1874), the Protrepticus, by G. Kaibel (Leips.

1894), the eiaayoiyrj öiaXeKTiK-f), by C. Kalbfleisch (Leips. 1896). J. Müller hasdiscussed the irepl AiroSei^eios.

P. 217. Line .3. Add :—

Of the new Berlin ed. of Philo, by L. Cohn and P. Weudland, Vols. I. and ILhave appeared (1896-1897).

P. 217. Line 14. To the lit. on Justin Martyr, add:—H. Veil (Strassburg, 1893).

Page 374: Windelband, History of Philosophy, Vol. 1

354 Appendix.

P. 217. Line 20 from foot. To the notice of Tertullian, add :.—

.

He was a partisan whose hot-headed fanaticism did not shrink from any para«

doxical consequence.

P. 217. Line 3 from foot. To the notice of Clement, add :—With iron will and tireless activity he united the peaceful and conciliatory

spirit of scientific culture, with which he sought to exercise an influence in thepassionate ecclesiastical controversies of his time.

P. 218. Line 15. To the notice of Plotinus, add :—A fine, noble nature, in whom the deep inwardising and spiritualising of life,

which was the most valuable result of ancient civilisation, found its best embodi-ment.

P. 218. Line 29. Add:—Porphyry's ElaayooyTj els ras KaT-qyoplas was usually known in the Middle Ages

by the title de quinque vocibus.

P. 224. Line 3. Add a foot-note :—

Similarly in the Epistle to the Hebrews, the relation of Jesus to the angelsis set forth in the manner in which it is presented by Philo.

P. 234. Line 3 from foot of text, add :—

This transition is also connected with the fact that in the Chris-

tian view the activity of consciousness just described was considered

less from the theoretical than from the practical standpoint. The

freedom of the will is here the central conception. The Oriental

Church fathers in part stood nearer the intellectualism of the Hel-

lenistic philosophy, or at least made concessions to it; on the other

hand, among the western teachers of the Church who were in closer

touch with Rome the will was most strongly emphasised in both

psychology and theology. Among the latter the tendency is domi-

nant to regard the spiritual or immaterial principle as passive and

determined by its object in so far as it is knowledge, but as active

and determining in so far as it is will.

P. 238. After line 6, insert the following paragraph :—In this connection the conception of the infinite underwent a

transformation which gave it a radically different value (cf. Jon.

Cohn, Geschichte des Unendlichkeitsproblems, Leips. 1896). The mind

of the Greeks, directed as it was upon measure and definite limita-

tion, had originally looked upon the infinite as the incomplete and

imperfect ; it was only with reluctance that when considering the

infinitude of space and time metaphysics had allowed itself to

ascribe to the infinite a second subordinate kind of reality, as was

•done by the Pythagoreans, the Atomists, and Plato— aside from

the isolated case of Anaximander, whose influence lay in another

direction. Now, infinitude had become the only predicate which

Page 375: Windelband, History of Philosophy, Vol. 1

Appendix. 355

could be ascribed to the highest reality or to the deity, as over

against the finite things of the world. Even the " negative " theology

could permit this expression. The name " infinite " must be applied

to the divine power which in the Stoic and Neo-Pythagorean phi-

losophy of nature was regarded as the essence pervading andinforming the world with its workings ; to the One from which

Neo-Platonism regarded worthy of the world's forms as flowing

forth; to the creative divine will which, according to Christian

teaching, had called forth the world from nothing, and thus shown

its freedom from all limitation; and finally to this supreme person-

ality himself in contrast with finite persons. Thus through this

final development of ancient philosophy the conception of the in-

finite became the constituent mark of the highest metaphysical

reality; it belongs not only to the universe as extended in space,

but also to the inmost essence of things, and, above all, to the deity.

This latter fusion became so fixed and sure that to-day it appears

entirely a matter of course in the sphere of thought, as well as in

that of feeling, to conceive of the supreme being as the Infinite, in

contrast with all finite things and relations.

P. 256. Line 11. To the phrase "drama of universal history"

affix the following foot-note :—

This expression has in this connection, as we see, a broader meaning, andone which conforms much more to the meaning of the words, than in its ordi-

nary use.

P. 263. To the literature of the period, add :—

B. Hauröau, Notices et Extraits de quelques Manuscripts de la Bibliotheque

Nationale. 6 vols., Paris, 1890-1893; H. Denifle and E. Chatelain, Chartula-

rium Universitatis Parisiensis. 2 vols, Paris, 1890-1894 ; H. Denifle and Fr..

Ehrle, Arch. f. Litt. u. Kirch. Gesch. d. Mittelalters, 1885 ff.

P. 273. Line 13. To the notice of Augustine, add :—

His youth was in part wild and irregular. His father, Patricius, belonged to

the old religion ; his mother, Monica, to Christianity. To a deeply passionate

nature he joined not only dialectical skill and keen intelligence, but aiso phil-

osophical subtlety and a wide intellectual and spiritual vision, which wasnarrowed only at the last by ecclesiastical partisanship. He was made bishop

391.

P. 274. Line 19.

"Eriugena" is given as first form of the name, with " Erigena" and " Jeru-

gena '

' as variants.

P. 274. Line 17, from foot, add :—

Recently his authorship lias been doubted and the work assigned to a Bern«hard Silvestris (also Bernhard of Tours).

P. 274. Line 14, from foot, add :—

Cf. A. Clerval, Les Ecoles de Chartas au Moyen-age (Chartres, 1895).

Page 376: Windelband, History of Philosophy, Vol. 1

356 Appendix.

P. 275. Line 5. To the notice of Abelard, add :—

The dialectical virtuosoship to which he owed his success and his fame de»

ceived both him and his time as to the slightness of his knowledge. On the

other hand, the freer and bolder convictions which he had gained in the ethical

and religious field by the keenness of his intellect could not overcome the coue-ter-tendency of his age, because they did not find sufficient support in his vain

and weak personality. In addition to the ed. in two vols, of his work, Cousinhas edited also Ouvrages inedits (Paris, 1886). Cf. S. M. Deutsch, P. A. ein

kritischer Theolog. des 12 Jahrhunderts (Leips. 1883); A. Hausrath, PeterAbälard (Leips. 1893).

P. 313. Line 25. To the lit. on the Amalricans, add :—

Cf. the Treatise against the Amalricans, ed. by CI. Bäumker (Jahrb. f. Philo;»

u. spec. Theol., VII., Paderborn, 1893).

P. 313. Line 15 from foot. To the lit. on Albert, add :—

V. Hertling, A. M. Beiträge zu seiner Würdigung (Coin, 1880).

P. 316. To the general lit. add :—

[T. J. de Boer, Gesch. d. Philos. in Islam (Stuttgart, 1901).]

P. 317. Add to third par. :—

Cf. T. de Boer, Die Widersprüche d. Philosophie nach Algazalli und ihr Au»gleich durch Ihn Boschd (Strassburg, 1894).

P. 320. Line 11, add: —But the " natural " man finds that even among a highly developed

people the pure teaching of the natural religion meets in most cases

only misunderstanding and disfavour. He turns back to his isola-

tion with the one friend whom he has gained (cf. Pocock's ed.

pp. 192 ff.).

P. 330. Line 3 from foot. To " Scotus," affix the reference :—

Cf. H. Siebeck, Die Willenslehre bei Duns Scotus u. seinen Nachfolgern,Zeitschrf. Philos. Vol. 112, pp. 179 ff.

P. 331. Line 9 from foot, add :—

It was a great service on the part of Buridan that, in order to

grasp the problem more exactly, he sought to state the question

once more in purely psychological terms. He sought to do justice

to the arguments on each side, and made it his purpose to develop

the conception of ethical freedom, in which indifferent!sm should

lose the element of arbitrary caprice, and determinism should lose

the character of natural necessity. Nevertheless, he did not succeed

in completely clearing up the complication of problems which inhere

in the word " freedom."

P. 333. Foot-note on word " synteresis," add :—

Cf., however, recently, H. Siebeck in Arch.f. Gesch. d. Philos., X. 520 ff.

P. 339. Foot-note 1. For " and the pseudo," read :—

*' and perhaps the pseudo,"

Page 377: Windelband, History of Philosophy, Vol. 1

Appendix. 357

P. 342. Line 24. Affix to « Occam," the reference :—Cf. H. Siebeck, Occam's Erkenntnisslehre in ihrer historischer Stelluna

Arch. f. Gesch. d. Philos., X. 317 ff.).

Page 378: Windelband, History of Philosophy, Vol. 1

INDEX TO VOLUME I

Abelard, life and writings, 274;

theory of universals (conceptual-

ism),"272, 294, 298 f.; rationalism

and independence, 300 f., 307; psy-

chology, 306 f.; ethics, 308 f.; reli-

gion, 319.

Absolute, applied to the Ideas, Plato,

128; to the pure Form, Aristotle,

145 f.; to the One, Plotinus, 238;

to God, Anselm, 293 f.

Abstract ideas, see Ideas.

Abubacer, 317, 320.

Academicians, 164.

Academy, Older, 101, 103, 159 (see

also under names of its adherents)

;

Middle, 103, 161 f., 207 (see also

Arcesilaus and Carneades) ; New,103, 162.

Acosmism, 38.

Actual vs. the potential, 140, 144, 146.

Adelard of Bath, 274, 297.

ddiä<popa in Stoicism, 168, 173.

iEgydius, 314.

iEnesidemus (the Sceptic), 160, 163;

his "tropes," 200; aporise, 206.

/Eons, Gnostic, 244, 257 f.

/Eschines, 82.

iEsthetics (see also Beauty), begin-ning of, in Aristotle, 153 ; Plotinus,

248.

Agrippa (the Sceptic), 160, 163; his

tropes, 201.

Alanus, 275.

Albert of Bollstädt (Albertus Mag-nus), 311, 313, 321, 326, 333, 340,343 f.

Aleidamas, 74.

Alcmajon, 46, 64, 67, 150.

Alcuin, 273.

Alexander Aphrodisias, 161, 234, 338 f.

Alexander of Hales, 313, 344.

Alexandria, 158, 213; Catechists,school of, 214, 217.

Alexandrian Philosophy, 213 ff.; seealso Neo-Pythagoreanism, Philo,Plotinus, etc.

Alexinus, 71, 89.

Alfarabi, 317.

Alfred de Sereshel, 344.

Algazel, 317.

Alhacen, 344.

Alkendi, 317.

Allegorical interpretation, 221 ff.

dWolwais and wepicpopd as kinds ofKivTjais, 39.

Amalric, Amalricans, 313, 339.

Amelius, 218.

Ammonius Saccus, 218.

Analytics of Aristotle, 104, 132-138.

dvafjivqcTcs (recollection), with Plato,

118, Augustine, 278.

Anaxagoras, life, 30 ; astronomical in-

terest, 41, 54; theory of elements,

41, 52; of the vovs, 41 f., 54, 62 f.,

185; influence of this on Plato,

128; and on Stoics, 187; teleology,

42, 54, 98 note; theory of cogni-

tion, 60, 62 f., 65; cf. 29, 91, 128,

185.

Anaximander, 27 ff ., 33 ff., 49, 60.

Anaximenes, 27, 29, 32 f., 48.

Andronicus, 104, 159.

Anniceris. 70, 87.

Anselm, 272, 295; life and writings,

274; ontological argument, 292 f.,

321, 331.

Antinomy, between thought and ex-

perience, 11; Zeno's antinomies,

44, 55 f.

Antiochus, 103, 161 f.

Antisthenes, 70, 72, 83 f., 94, 96; see

also Cynics.Apathy, Stoic doctrine of, 168.

äireipov, see Infinite.

Apelles, 258.

Apollodorus, 162.

Apollonius, 213, 215.

Apologists, 214, 217, 222 ff., 231, 237.

A posteriori, see A priori.

A priori, evolutionary explanationof. 105 ff.; 292 f., 343 ff.

Apuleius, 213, 216, 228.

Arabian Philosophy, 15, 316 f., 319,

337 ff.

Arcesilaus, 103, 160 f.

Archelaus, 76.

Archytas, 31, 103, 123, 215.

Aristarchus, 162.

Aristides, 217.

Aristippus, 70, 72, 85 ff., 93, 165, 170;

see also Cyrenaics.Aristippus the Younger, 70, 72, 86.

Aristobulus, 216, 220 f.

Aristophanes, 81.

Aristotelianism (see Peripatetics), in

Middle Ages, 269 f, 288, 302 f.,

311 ff., 316 f., 324 ff., 329, 333, 338.

Aristotle, conception of philosophy,

2; completer of Greek science, 25,

Page 379: Windelband, History of Philosophy, Vol. 1

Index.

99 f.; on 6avnäi;eiv and äpxv, 31 f.;

as source for Sophistic doctrine,8S; life and writings, 103 f.; logic,

132-138; his central principle, 139;doctrine of cause, 141 ff.; catego-ries, 142; relation to Plato's Ideas,

139, 142 f.; doctrine of matter,144; of Being or essence, 139 f.,

145 f.; monotheism, 145 f.; cos-mology, 147; psychology, 149;ethics, 151 ff.; politics, 152 f.; po-etics, 153 f.; influence on Stoics,

176, 181; immanence and tran-scendence in his doctrine, 178 f.;

on freedom, 191 f.; on law in na-ture, 195; evil due to matter, 196;influence of his monotheism, 211;reception of his doctrine the deci-

sive factor in Scholasticism, 269,311 f.; cf. also 229, 236, 255, 320,

331, 340; see also Aristotelianism.Aristoxenus, 159, 161.

Arius Didymus, 162, 216.

Arnobius, 214, 217, 224 f.

Arrian, 216.

äpxv of cosmologists, 32 ff.; theIdeas as äpxv with Plato, 118; fourprinciples, Aristotle, 138, 141.

Asceticism, 230.

Aseity, of God, 292.

Assent, as characteristic of thejudgment, 207; as ethical factor,

308.

Astronomy, of the Pythagoreans, 45,

56 f.; of Anaxagoras, 54; of Plato,

130 f.; of Aristotle, 147 f.

Ataraxy, 165; of Epicurus, 166; of

Sceptics, 167; of Stoics, 168.

Atheism, 86.

Athenagoras, 217, 224.

Atom, conception of, with Leucip-pus, 43; of Democritus, 107, 110 ff.;

with Epicurus, 184.

Atomism, of Leucippus, 42; of

Democritus, 108, 110 ff.; of Epicu-rus, 183 f.; in Ethics, see Individ-

ualism.

Atomists, 29, 42 ff., 54; see also

Leucippus, Democritus.Augustine, 264 ff., 268, 270; life and

works, 273; doctrine, 276-287; in-

fluence of his theory of the will,

311 f., 329 ff.; his emphasis onpersonality and inner experience,

303, 340, 344; influence on Re-formers, 337; cf. also 324, 326,

333. 337, and Augustinianism.Augustinianism, contrasted with

Aristotelianism, 303 ff., 324, 326,

329 ff., 334, 341, 344.

Authority as philosophical principle,219 ff.

Avempace, 317.

Averroes and Averroism, 317 ff., 320,323, 329, 331, 336, 338 ff.

Avicebron, 318, 332, 338 f., 341.Avicenna, 299, 317, 340, 344.

Bacon, Roger, 314, 319, 333, 341,344 f., 307.

Bardesanes, 217, 239.

Basileides, 214, 217, 243, 258 f.

Beauty, its relation to the good withPlato, 125; first treated indepen-dently by Plotinus, 248 f.; Cf.^Esthetics.

Becoming; see Cosmic processes.Bede, 273.

Being, early Greek conceptions of,

31-47; as world-stuff with Miles-ians, 32; as corporeality or space-filling substance, Parmenides, 37;plurality of, assumed, 39 ff.;

= atoms, 42 f.; plurality of, deniedby Zeno, 44; found in numbers,Philolaus, 45; identified with thegood by Euclid, 96; equivalent toatoms with Democritus, 108; toIdeas with Plato, 109, 118; toessence with Aristotle, 139; andfurther to pure thought, 145; tospirit with Neo-PIatonism andPatristic thought, 232; with Ploti-

nus, 245; sought in the universalby John Scotus, 289 ff.; treated asan attribute of varying intensity,

291 f.

Berengar, 275, 297.

Bernard of Chartres, 272, 274, 294,302 f.

Bernard of Clairvaux, 273, 275, 301,305.

Bias, 24.

Bodies, as portions of space, Pythag-oreans, 46 f.; Plato, 129.

Bodv and Soul, 301 f.; see Soul.Boe'thius, 270, 273, 288, 296.

Bonaventura, 313, 333 f., 341.

Boyle, 380.

Brucker, 10.

Buridan, 315, 331.

Cabbala, 317.

Callicles, 75.

Callippus, 147.

Carneades, 103, 160 f., 194 f., 201,

207.

Carpocrates, 217, 258.

Cassiodorus, 270.

Catch Questions among the Sophistsand Megarians, 89.

Page 380: Windelband, History of Philosophy, Vol. 1

Index.

Categories, Aristotle's, 142; Stoics,

198 f.; of Plotinus, 245; natural

categories not to be applied to

God, according to Augustine,

279 f.

Cause and causality, Idea as, withPlato, 128; four causes of Aristotle,

141; final and mechanical, 144;

emphasized by Stoics, 181; con-

ception of, criticized by Sceptics,

205 f.

Celsus, 216.

Cerdo, 258.

Cerinthus, 257.

Chance and contingent, with Aris-

totle, 143, 148.

Change, as problem of philosophy,

47 ff.; law of, with Heraclitus, 50;

denied by Parmenides, 51.

Christianity, relation to Greekthought, 212, 223 f.; its view of

authority and revelation, 221 ff.;

of spirit and matter, 231 ff.; of

personality of God, 238, 251; its

view of history, 256 ff.

Chrvsippus, 159, 162, 168, 181, 187,

193 f., 196, 203.

Church, conceived as fellowship,

261 ; Thomas, Dante, Occam, 326-

328 ; attitude toward Aristotle, 312,

364; and state, theories of, 326;

preserves ancient civilisation andeducates modern Europe, 263 ff.;

one of the foci of Augustine's

thought, 276, 283.

Cicero, 161 f., 163, 177, 204, 223.

Civilisation, as factor in history of

philosophy, 13; influence on an-

thropological period of Greekthought, 66 ff.; its worth denied

by Cynics, 84; affirmed by Cyre-

naics, 86; the Hellenistic, 155 ff.;

preserved by the Church, 263 ff.

Civitas dei, of Augustine, 285.

Cleanthes, 159, 162, 188.

Cleidemus, 70.

Clement of Alexandria, 214, 217, 252.

Clement of Rome, 259.

Clitomachus, 161.

Coincidentia oppositorum, of Nicol-

aus Cusanus, 346.

Conception, its importance withSocrates, 95 f.; relation to Ideawith Plato, 118 f., 121; with Aris-

totle, 133, 142 f.; derived fromsense perception by Stoics andEpicureans, 203; Abelard's theory,

306.

Concepts, Aristotle's doctrine of,

137; Occam, 342 f.

Conceptualism, 272; of Abelard, 298.

Conscience, 234; Abelard's view of,

308; Thomas, 333; as synteresis,333.

Consciousness, denned, 234; as aunitary function with Aristotle,

150; characteristic of man, withAlcmaeon, 64 note 4; certainty of,

as starting-point with Augustine,276 f.

Consensus gentium, 204.

Conservation, of energy, 37-39.

Constantinus, 302.

Contemplation, 306; sethetic, 250;intellectual, 154, 286, 333.

Contingency of the finite, 347; in

freedom of the will, 330; of theindividual, 341.

Contract theory of the state, 174 f.,

328; see also state.

Contradiction, principle of, 61, 88,

138.

Copula, 37.

Cornutus, 216.

Cosmic Processes, early Greek con-ceptions of, 47 ff.; Aristotle's

principle for explaining, 140, 144;

see Change.Cosmogony, poetic, 27; emanistic,

249; earlv physical, 47 ff.

Cosmological argument, 145.

Cosmopolitanism, Stoic and Roman,176 f.

Crantor, 103, 164.

Crates of Athens, 103.

Crates of Thebes, 72, 85.

Cratylus. 70.

Creation, opposed to evolution andemanation, 252-254.

Criteria, of truth, 197 ff. ; see also

Rationalism and Empiricism of

true revelation, 225 f.

Critias, 76.

Criticism, immanent, 18.

Cusanus, see Nicolaus.

Cynics, 70, 82 ff., 90, 94, 96, 164, 166.

169, 171.

Cyrenaics, 70, 82, 86 f., 94, 165.

Daimonion (or Daemon) of Socrates,

98.

Damascius, 215, 218.

Dante, 311, 314, 327, 334.

Darwinism, with Empedocles, 53; see

Natural Selection and Survival of

the fittest.

David of Dinant, 313.

Deduction, Aristotle's conception of,

134.

Definition, Socrates, 95; Aristotle,

137 f.

Degerando, 10.

Page 381: Windelband, History of Philosophy, Vol. 1

Index.

Deity, first used as philosophicalprinciple by Anaximander, 34; asIdea of the Good, Plato, 128; asdemiurge, Plato, 130; as pureForm, with Aristotle, 145; aspneuma, with Stoics, 186 f

.; Epicu-

rus' view of, 188; knowledge andBeing, 335; distinguished fromGod, 335; as natura naturans,with Eckhart, 335 f.; see also God.

Demetrius, 216.

Demiurge, Plato's idea of, 130;Valentinus, 254; Gnostics, 257 ff.

Democritus, belongs to SystematicPeriod, 25 f., 99 f.; life andwritings, 100 f.; grounds meta-physics anew, 105-108; his systemof materialism, 109-116; relation

to Plato, 105-108, 118 f., 130; toAristotle, 138 f., 148 ff.; to Epicu-rus, 165, 183-185, 202; to Stoics,

180 f.

Demonax, 213, 216.

Determinism, Socrates, 79 f.; Stoics,

193 f.; opposed by Carneades andEpicurus, 194 f.; intellectualistic,

330; see also Freedom.Development, Aristotle's central

principle, 139 ff.; Thomas, 324.

Dexippus, 218.

Diagoras, 76.

Dialectic, of Zeno, 44, 55 f.; of Soph-ists, 69, 88 ff.; of Plato, 120; of

Aristotle, 132 f., 137; of Proclus,

251; of Scholasticism, 271; op-posed by the Mystics, 272; of

Abelard, 300.

Dicsearchus, 159, 161.

Didymus, see Arius.

Diodorus Cronus, 71, 89.

Diogenes Laertius, 215.

Diogenes of Apollonia, 32, 55, 62 ff.,

70, 150, 187.

Diogenes of Babvlon, 162.

Diogenes of Sinope, 70, 72, 84 f., 94.

Dionysidorus, 89.

Dionysius the Areopagite, 271, 274.

Docta iqnorantia, with NicolausCusanus, 337, 343, 347.

Dominicans, 313, 340.

Doubt, as Augustine's starting-point,

277.

öö£a, see Opinion.Dualism, of Pythagoreans, 46; of

Plato, 120, 130; overcome by Aris-

totle, 133; ethical and religious, in

Alexandrian thought, 229 If., 235ff.; of Gnostics and Maniohreans,239 f.; with Augustine, 285 f.;

anthropological, of body and soul,

304 ff.

Svvapis, 140, 179; see also Potential,Power, Dynamic.

Duns Scotus, personality and writ-ings, 311, 314; separates ihcologyfrom philosophy, 322 f.; meta-physical psychology, 324 f.; in-

determinism, 330, 332 f.; onrelation of intellect and will, 334,

on individuality, 341 f.; gave im-petus to empirical science, 344.

Duty, Stoics, 172.

Dynamic conception, Strato and the

Stoics, 179, 236.

Eckhart, 311, 314, 330, 332, 334 ff,

340.

Eclecticism, ancient, 161.

Ecphantus, 46, 56.

Ecstasy, with Philo, 227; Neo-Platonism, 228 f., 250.

Education, in Plato's Republic, 127;of the human race through revela-

tion, 226.

ffyenoviKov, 172, 179, 187, 339.

eiduXa, 113-115, 188.

Elean-Eretrian School, 70, 82.

Eleatics, 28, 30, 34 ff, 51 ff, 59 ff,

89 f.; see also Xenophanes.Elements, of Empedocles, 39 f.; as

homoiomeriai, with Anaxagoras,

41 ; of Pythagoreans, 57 ; withAristotle, 147 f.

Emanation, in Alexandrianism, 242f.; as eternal necessity, 249; as a

logical system, 250 f.; withErigena, 289-291.

Emotions, ancient conception of,

165; Stoics on, 168.

Empedocles, 29 f, 39 f, 51 ff, 58 ff,

92.

Empiricism, favoured by Nominal-ism, 344.

iv Kal wdv, 35,; of. Pantheism.End, see Teleology.ivepyeia, 140, 144.

Encrev. specific of the sense organs,

65, 113.

Enlightenment, the Greek, 66 ff.

Ens realissiynum, et perfectissimum,

292.

Entelechy, of Aristotle, 140 ff.

Epicharmus, 66.

Epictetus, 213, 216, 230.

Epicurus and Epicureanism, 158 f.;

life and writings, 162; ethics andtheory of life, 165 f, 170 f.; theory

of the state, 173 ff, 328; view of

Nature, 180, 182-1S6, indetermin-

ism, 193 f.; logic and theory of

knowledge, 198, 202 f., 205; cf.

also 211 f, 229, 252.

Page 382: Windelband, History of Philosophy, Vol. 1

Index.

Epistemology, or theory of cogni-tion, origin of its problems, 58; of

Greek cosmologists, 58-65; treated

psychologically by Protagoras, 91

ff.; of Aristippus, 93 f.; of Socrates,

94 ff.; made basis of meta-physics, 101, 104 ff.; of Democri-tus, 104 ff., 110 ff.; of Plato, 104

ff.. 117 ff.; the principle of Aris-

totle's logic, 133; Stoic, 199, 207ff.; of Sceptics, 200-202, 205-207;

of Epicureans, 204 f.; of Augus-tine, 277-282; of Occam, 325; of

Mysticism, 335 ff.

Eratosthenes, 162.

Eric of Aux., 273.

Erigena, John Scotus, 271, 274, 289-

291, 335.

Esse, in intellectu and in re, 293;with formaliter, 325; with nosseand velle, Augustine, 280.

Essence, with Aristotle, 139, 141 andexistence, 293 ff.

Essenes, sect of, 213, 231.r

Eternal truths, see Verites.

Eternity, of the world, Aristotle. 144f.; Origen, 253 f.; Plotinus, 249;and time, 287.

Ethics, principle of, first propoundedby Heraclitus, 63 ;

problems raised

by Sophists, 72 ff.; intellectual-

istic and eudaemonistic of Socrates,

77 ff.; of Democritus, 115 f.; of

Plato, 123 ff.; the basis of his

idealism, 108 f., 117 f.; of Aristotle,

151 ff.; of the Stoics, 163 ff.; of

Epicureans, 165 ff.; of Sceptics,

165 ff.

; of Augustine, 287 ; of Abe-lard, 308; of Thomas, 332 f.

Eubulides, 71, 89.

Euclid, 70 f., 89, 96, 102.

Eudsemonism, in Greek ethics, 79ff., 87, 151.

Eudemus, 161, 198.

Eudorus, 216.

Eudoxus, 103, 147, 186.

Euemerus, 70.

Euripides, 66.

Eusebius, 216.

Euthydemus, 89.

Evil (see also Theodicy) in theworld, 195-187; negative withPlotinus and = matter, 247; Pa-tristic doctrine of, 252 f.; negativewith Augustine, 280.

Evolution, as opposed to emanation,243; see also Development, Nat-ura Selection etc.

Existence; see Essence, and also

God.Experience, in opposition to thought,

58 f.; Democritus and Plato, 105

f., 110, 114 f.; cf. see Empiricism.

Peeling, with Cyrenaics, 86; Vic-torines, 305.

Figulus, 215.

Final causes; see Cause.Eire, as first principle, Heraclitus, 36,

50.

Fischer, K, 13.

Force, moving, Empedocles, 40; An-axagoras, 41; see also Conserva-tion.

Foreknowledge of God, as argumentof determinism, 193.

Form, essential nature of things,

with Democritus, 107, 111 ff.; withPlato, 107-109, 129 (see also Idea)

;

contrasted with matter by Aris-totle, 139 ff.; pure, 144 f.; in

psychology of Scholastics, 324 f.;

with Averroes, 338; individualForms with Scotus, 341.

Francis of Mayro, 315, 342.

Fredegisus, 274.

Frederick II. of Sicily, 319.

Freedom, ethical, maintained bySocrates, 191 ; distinguished fromfreedom of choice by Plato, 191

;

Aristotle's conception of freedom,192; Stoics' deterministic views,

193 ; metaphysical freedom as in-

determinism of Epicurus, 194 f.;

applied to God by Patristic

thought, 252; used to explain

origin of evil, 252 f.; both main-tained and denied by Augustine,282-285; maintained as determin-ism by Thomism, 329 f.; as inde-

terminism by Scotus and Occam,330 f.; as ethical, Buridan, 331.

Fulbert, 302.

Galen, 216, 316.

ya\7ipio-fi6s, 166; cf. 116.

Gaunilo, 293.

yiveais, with Plato, 106, 120.

Gerbert, 272, 275, 302.

Gerson. 315, 323.

Gersonides, 318.

Gilbert, 275, 335.

Gnostics, 214, 217, 222, 224, 237, 239,

243, 257.

God (see also Deity, Theology), first

philos, conception of, as matter,Anaximander, 34; as ev ko.1 irdv,

Xenophanes, 34 f.; his relation to

the world in Hellenistic thought,235 ff.; exalted above all mind or

matter, 237 (see "Negative Theol-ogy") ,* personality of, in Chris-

tainity, 238, 251; personality of,

rejected by Greek and Neo-Platonic thought, 238; implicit

Page 383: Windelband, History of Philosophy, Vol. 1

Index.

and explicit, 290, 346; sourceof truth for Augustine, 278 f.;

Anselm's argument for existenceof, 292 f., 485 ; distinguished fromdeity by Gilbert, 335.

Good, the, Socrates leaves it

undefined, 79; virtue with Antis-thenes, S3; pleasure with Aristip-

pus, 85; Idea of, with Plato, 122 f.,

125; happiness or well-being withAristotle, 151 ;

pleasure withEpicurus, 165 f., 170; virtue withStoics. 168 ; absorption in the deitywith Neo-Platonists, 250; contem-plation with Augustine, 286 f.;

and Thomas, 333 f.; love withScotus, 334; intellectual love ofGod with Spinoza, 435; recog-nized by God's wisdom, ace. toThomas, 332 ; result of God's will,

Scotus, 332.

Gorgias, 30, 69, 71, 89 f.

Gottfried of Fontaine, 330.

Grace, realm of, opposed to nature,318 ff.; irresistible with Augustine,282, 284; supported by Thomas,denied by Scotus, 334.

Grammar, blended with logic withthe Sophists, 88, 96; Terminists,342 f.

Gregory of Nyssa, 254, 261.

Grote, 71.

Harmony, of the world, according toHeraclitus, 36, 49 f.; and spheres,

Pythagoreans, 45.

Heaven, as realm of order and per-fection, according to Anaxagoras,41 f., 54; Pythagoreans, 57; Aris-totle, 147.

Hedonism and Hedonists, 70, 85 ff.;

93 f.; of Epicurus, 165 f., 170 f.;

of Pyrrho, 167; see also Syrenaics,

Epicureanism, Ethics, Utilitarian-

ism and Eudsemonism.Hegel, conception of history of

philosophy, 10 f., 13.

Hegesias, 70. 87.

Hellenistic philosophy, 155 ff.

Hpnads, 251.

Henrv of Ghent, 314, 330 f., 340 f.,

345.

Heracleides Lembus, 161.

Heracleides of Pontus, 103.

Heraclitus, general character of his

thought, 28; life and writings, 30;

as a reformer; conception of theuniverse, 36 ff . : of the cosmicprocess, 49 f.; of cognition. 58 f.;

influence on Protagoras, 92; onthe Stoics, 186, 209; on ^Eneside-mus, 200; cf. also, 72, 118.

Herennius (Pseudo.), 277.Hermes (Trismegistus), 216.Hermetic writings, 227, 237.Hermippus, 161.

Hicetus, 56.

Hieroclcs, 218.

Hildebert of Lavardin, 275.Hippasus, 57.

Hippius, 69, 71, 73 f., 88.Hippo, 70.

Hippodamus, 66, 74.

Hippocrates, 67, 316.Hippolytus, 214, 217.

History, philosophy of, 19 ; its worthfirst recognised by Cicero, 177;problem of, first suggested byChristianity, 255 ff.; Patristicviews of, 256 ff.; with Augustine,285 f.

History of Philosophy, see Philos-ophy.

Homoiomeriai of Anaxagoras. 41.

Hugo of St. Victor, 275, 305, 324, 334.

Hylozoism, of the Milesians, 32, 44,

48; dynamic, with Strato, 179.

Ibn Tofail, see Abubacer.Idealism, Plato's system of, 116-131;

psychological and epistemologicalof Occam, 325 f.; Neo-Platonic ofEckhart, 335; see also Ideas.

Ideas, (1) In Platonic or relatedsenses: with Plato, 109, 118 ff.;

Aristotle's criticism on Plato'sIdeas, 133; their influence onhim, 142 f.; Plato's theory op-posed by Stoics and Epicureans,203; innate, 204; Plato's Ideastransformed to thoughts of Godby Neo-Pythagoreanism and Neo-Platonism, 233; Philo's doctrine,240 f.; Plotinus, 245; Augustine,279; in mediseval thought (see

Universals) ; see also Idealism,Neo-Platonism. Conception.

— , (2) In sense of a mental modifica-tion or content (Ger. Vorstel-

lung) : transition from Platonicusage, 203. 306.

Identity, principle of, exaggeratedwith Sophists, 89 f.; of thoughtand beine with Parmenides, 37 f.

Image, 113-115, 1SS.

Imagination, 2S1, 306 f.

Imitation, as essence of art, 153 f.

Immanence and transcendence of

God, 178 f., 235 ff., 212 ff., 245. 337f.

Immatorialism, Plato's. 109. 116 ff.

Immortality of the soul, in myth, 62note 1 ; problematic with Socrates.

79; asserted with Plato, 124; and

Page 384: Windelband, History of Philosophy, Vol. 1

Index.

in Platonism, 232; with Aristotle,

150 f.; Stoics, 187; lost in pan-psychism, 339 f. ; not demonstrableaccording to Duns and Occam, 322.

Imperfection, see Evil and Theodicy.Indeterminism, 194 f., 329 ff.; seeFreedom.

Indifferentism, 297.

Individualism, of Democritus, 116;of Greek epigones, 163 ff. ; of

Epicurus, 170 f.

Individuality, problem of, 337 ff.

Induction with Socrates, 97; Aris-

totle on, 137.

Infinite, regarded by Greeks as im-perfect, 46; the, of Anaximander,33; contradictions involved in

conception of, Zeno, 44; attribute

of the deity, Neo-Platonism, 236;of divine will, Origen, 252; of

God with Cusanus, 345-347.

Innateness, of ideas, Cicero andEclectics, 204.

Intellect, its relation to will withThomists and Scotists, 329 ff.;

active and passive with AlexanderAphrodisias and Averroes, 339 f.;

see also Reason, Nous, Will, In-

tellectualism.

Intellectualism, of early science, 62;of Socrates, 79 f.; of Democritus,115 f.; of Aristotle, 151, 154; of

Augustine, 286 f.; of Thomas,330, 333 f.; of Eckhart, 334-337.

Intuitive knowledge, with Plato,

118 f.; Occam. 342 f.

Irenaeus, 217, 221 f., 224, 226, 232,

259, 261.

Irony, with Socrates, 97.

Isidore of Sevilla, 270, 273.

Jamblichus, 31, 215, 218, 220, 222,

226, 250.

Jesus, his influence, 223; as centre

of world's history, 256 ff.

Jewish philosophy, 317.

Joachim of Floris, 319.

John of Brescia, 320.

John of Damascus, 271, 273.

John of Rochelle, 344.

John of Salisbury, 276, 307.

Judgment, Aristotle's treatment of,

135 ff.; Stoics, 207 f.; with Augus-tine, 278-280; see also Logic.

Julian, 218.

Jus naturale, 111; see Law, andRight.

Justice, as principle of the state,

with Plato, 127.

Justin Martyr, 214, 217, 223 f., 237,

259.

Kalokagathia, Socrates, 79.

Kant, conception of philosophy, 4.

Kdöapais, Aristotle's doctrine of,

153 f.

Knowledge, as participation in worldconsciousness, 63 f.; as copy of

reality, 114, 119, 202, 325; cf. alsoSigns; as recollection, 118 ff., 223;as impersonal and super-personalfunction, 339 f.; limits of, withSocrates, 97 f.; as end in itself,

23, 350; as set over against faith,

322 f.

Lactantius, 217.

Lanfranc, 275.

Language, bearing on philos. studies

by Sophists, 87 f., 96; by Abelard,306.

Law, first grasped clearly by Hera-clitus, 37, 50; suggested by mathe-matics and astronomy, Pythagor-eans, 56 f.; relation to Nature, 73;emphasised by Democritus, 111;by Stoics, 181.

Leucippus, 29 f., 42 f., 52 ff., 60, 108,

111, 128 f.

Lewes, 11.

Life, as principle of explanation withIonics, 32; with Aristotle, 141.

Logic, defined, 20; Sophists, 88 ff.;

Socrates, 97 f.; Plato's, or dialec-

tic, 119 ff.; Aristotle's, 132-138;

Peripatetics, 197 f.; Stoics, 198 f.;

hypostatisation of logical processes

by Porphyry and Proclus, 250 f.;

main topic of Middle Ages, 270 f.

;

logical relations identified withmetaphysical, 290; formal logic

the only possible for empiricism,

360 f.; of Ramus, 361; terministic

of Occam, 342; see also Dialectic

Realism.Logos, doctrine of, with Heraclitus,

36 f.; Stoics, 180 f., 186; influence

of Stoic doctrine on Christian,

223 f.; Philo's doctrine of, 241 f.;

Origen, 254.

Longinus, 218, 233.

Lucretius, 162.

Lullus, 315, 321.

Lyceum, see Peripatetic School.

Lycophron, 74 f.

Macrocosm and microcosm, 187.

Maieutic, 97.

Maimonides, 318 f., 321.

Man, identified with animal world,

52 f.

Mani and Manichseism, 239 f., 286.

Marcianus Capella, 273, 296.

Page 385: Windelband, History of Philosophy, Vol. 1

Index.

Marcion, 221, 258.

Marcus Aurelius, 213, 216, 230.Marsilius of Inghen, 315.

Marsilius of Padua, 345.Materialism, of Leucippus, 43; ofDemocritus, 108, 109 ff.; of Epicu-reans, 183-186; of Stoics, 186.

Mathematics, with Pythagoreans, 45-

47, 56 f.; in Plato's system, 129.

Matter, cosmic, of Ionics, 32; Anaxi-mander, 33; opposed to form byAristotle, 139 ff.; accessory cause,144; Non-being or space withPlotinus, 246 f.; evil, 247; regardedas self-moved, etc., by Averroes,338.

Maximus Conf., 274.

Maximus of Tyre, 216.

Mechanism and mechanical view ofworld, Leucippus, 53; with Strato,179; Epicurus, 183.

Medicine, independent origin, 2;serological, 66.

Megarians, 70 f., 82, 89.

Melissus, 28, 30, 44.

Melito, 217.

Menedemus, 72.

Metaphysics, origin of name, 19;grounded anew by Democritus andPlato, 104; Plato's teleological,

128; connected with logic, 133; of

Aristotle, 139 ff.; of Theophrastus,178; of Stoics, 180; religious, 214ff.; of logical genera and species,

271 ff.; of inner experience, 276ff.; logical, of Realism, 290 ff.; ofNominalism, 296; of psychology,323 f.

Method, maieutic of Socrates, 97;modified by Plato, 118 f.; Aris-totle's deductive, 137 ff.; scholas-tic, 312, 344; inductive, 97, 118,

137, 344.

Metrodorus, 76.

Metrodorus the Epicurean, 162.

Michael Psellos, 342.

Microcosm, see Macrocosm.Milesians, 28 f., 32 ff ., 48 ff

.

fiißtjffLs, 47, 120.

Minucius Felix, 214, 217, 224.

Moderatus, 215.

Monism, original presupposition, 32ff.; metaphysical, of the Eleatics,

37 ff.; of the spirit, in Neo-Platonism, 240 ff.

Monotheism, pantheistic with Xeno-phanes, 34; of Cynics, 85; theistic

with Aristotle, 145 f.

Morals, Plato's, 125 ff.; ascetic, 230.

Motekallemin, 317.

Motion, as basis of mediating at-

tempts, 39; the essence of change,

43; early theories of its cause, 52

ff.; contradictions in conceptionof, Zeno, 55; basis of feelings withCyrcnaics, 86; of perceptions withProtagoras, 92; with Democritus,113 f., 115 f.; with Aristotle, 147 f.

Motives, Greek theories, 72, 75, 79 f.

Music, theory of Pythagoreans, 45.

Musonius, 216.

Mutazilin, 318.

Mysteries, 124.

Mystics and Mysticism, source in

Neo-Platonism, 227; a factor ofMed. philos, 266 ff., 275, 304 ft'.,

333.

Myths, with the Sophists, 76 ; Plato,

102, 123, 687 (123); Stoics, 189 f.;

Gnostics, 243 f.

Naturalism of Strato, 179; of Arab-ians, 338; see also Materialism,Mechanism.

Natural law, see Law, and Right.Natural selection, 53.

Natural science, among the Greeks,27 ff.; daughter of Humanism, 351

;

favoured by Nominalism, 343 f.

Natura Naturans and Natura Na-turata, probably first used byAverroism, 336, 338; with Eckhart,335 f.

Nature, first object of philosophy, 25,

27 f.; contrasted with statute, 73

ff.; with Democritus, 116; Plato's

philos. of, 129 f.; Aristotle's, 146 ff.;

Stoic doctrine of life according to,

171 f.; regarded as equivalent to

law, 171 ; Strato's view of, 179

;

Epicureans' view of, 183 ff. ; Stoics',

186 f.

; spiritualisation of, by Ploti-

nus, 249; by Valentinus, 254; re-

turn to, by school of Chartres, 302

f.; relation to deity with Eckhart,

335.

Nausiphanes, 165.

Necessitv, mechanical, with Leucip-pus, 53; with Plato, 130; logical,

with Aristotle, 134; natural, withStoics, 181; denied by Epicurus,183.

Negative theology, with Philo, Apol-ogists, and Neo-Platonists, 237 t'.,

with Scotus Erigena, 290; of Eck-hart, 335.

Nekkam, Alex, 344.

Neo-Platonism, dependent on earlier

Greek conceptions, 123, 157; per-

sonality and writings, 215, 21S;philosophical interpretation of

myths, 222; on spirit and matter,233 f.; doctrine of Ideas, 117 note6: 233 note 2; on nature of God.237 ff.; on history, 255; in MiddleAges, 268 ff.; influence on Angus-

Page 386: Windelband, History of Philosophy, Vol. 1

Index.

tine, 279 f., 286; on John Scotus,

289 ff.; on Bernard of Chartres,

294; on William of Champeaux,295; see also Plotinus, Proclus.

Neo-Pythagoreans, 117 note 6, 123,

213, 215, 220 f., 230 f., 233, 237.

Nicolas d'Oresme, 345.

Nicolaus d'Autricuria, 344.

Nicolaus Cusanus, 312, 315, 335 f.,

337, 343, 345 f.

Nicomachus, 213, 216.

Nigidius Figulus, 215.

Nominalism, 272; its origin, 296; of

Roscellinus, 296 f.; revived, 312,

342; favours study of natural sci-

ence, 343 f.; see also Terminism.Norms, 63, 69, 181, 279.

vovs, of Anaxagoras, 42, 54, 63; as

part of soul with Plato, 124; withAristotle, 150; with Theophrastus,178 f.; Plotinus, 245; Augustine,

279, note 3 ; see Reason.Numbers, with Pythagoreans, 45, 47

;

with Plato, 122, 129, 131 ; in Alex-andrianism, 242 ff.

Numenius, 213, 216, 220, 223, 232.

Occam, see William of Occam.Odo (Odardus) of Cambray, 295.

Oinomaos, 216.

One (eV), of Xenophanes, 34 f.; withParmenides, 38; with Neo-Pythag-oreans, 237 f.; with Plato, 122;with Proclus, 251.

Ontological argument of Anselm,292 f.

Ontology, of the Stoics, 199.

Ophites, 258.

Opinion, opposed to knowledge, 58,

95, 105-117; to sense perceptions,

204; relativity of, 201.

Optimism, religious, 252.

Order, Heraclitus, 36, 49; as norm,63; Anaxagoras, 42, 54.

Organism, as principle with Aristotle,

141.

Organon, of Aristotle, 104, 132 ff.

Orient, its philosophy, 23 note; influ-

ence, on Greeks, 27. 211, 213 ff.; onMiddle Ages, 310, 316 ff.

Origen, the Christian, 214, 216 ff ., 222,

233, 235, 253 f., 261.

Origen, the Neo-Platonist, 218.ovcrla, with Plato, 106 ff., 120-123;

Aristotle, 139 ff.; Plotinus, 245;Origen, 254.

Pain, see Pleasure.Pansetius, 161 f., 190.

Pan-psychism, 340.

Pantheism, suggestions for in Eleati-cism, 34 f., 37; Strato 's, 179; ofStoics, 180; in conjunction with

theism, 236; logical of Realism,295; of Averroism, 313, 338 ff.; ofAmalricans, 339.

Parmenides, 28 ff., 37 ff., 46, 51, 58 ff.,

90, 118, 129 f.

irapovffia, 120.

Participation, of things in the Ideaswith Plato, 120.

Particular, see Universal.Passions, ancient conception of, 165;

Stoics on, 168; cf. Emotions.Patristics, 214.

Perception, contrasted with reflective

thought by cosmologists, 58 ff.;

Protagoras's theory of, 91 ff.; De-mocritus, 105, 113 ff. ; Epicureantheory, 202; Stoics', 202.

Peregrinus Proteus, 216.

Peripatetic School, 103, 159, 161, 164,

178, 180, 229; see also Aristote-lianism.

Perseitas boni, 332.

Persius, 216.

Personality, emphasised in Hellen-istic thought, 223; found in spirit,

232; Christian view of, 251; em-phasised by Christian thinkers asagainst Arabian pan-psychism, 340.

Pessimism, among the Cyrenaics, 87;among Stoics, 169; in Christiandoctrine, 252.

Peter Lombard, 275.

Peter of Poitiers, 275.

Petrus Aureolus, 315.

Petrus Hispanus, 315, 342.Phaedo, 72.

Phaedrus, 162.

Phaleas, 74.

Phenomena and phenomenal, objectsof sensation or perception as, 92 f.,

106, 110; cf. also Supersensuous,Rationalism, Knowledge.

Pherecydes, 24, 34.

Philip, or Philippus, of Opus, 103, 123.

Philo of Larissa, 103, 161 f.

Philo of Alexandria, 214, 216, 220 ff.,

227, 231, 237, 240 ff., 290, 319.

Philodemus, 162, 198, 342.

Philolaus, 29, 31, 45, 60 f., 63, 129, 215.

Philosophy, various conceptions of,

1 ff. ; relation to other sciences, 5;to civilisation, 6, 13 ; external posi-tion, 7; share of different peoplesin, 8; division of, 18 ff.; sources of,

among the Greeks, 23 ff., 27 ff.; atfirst cosmological, 27 f.; then an-thropological and practical, 68 ff.;

Aristotle's division of, 153; sepa-rating of special sciences from, 156;as wisdom for life, 157 ff. ; fusedwith religion, 210 ff.; relation toChristianity, 224 ff.; to theologywith Scholastics, 321.

Page 387: Windelband, History of Philosophy, Vol. 1

Index.

Philosophy, history of, defined, 9;Hegel's view of, 10 f., 12 f., 614,

681 ; Fischer's view of, 13 ; threefactors in, 11-14; tasks of, 15;sources for, 15 ff.; its significance,

681 ; and see also each of the pe-riods and writers treated ; divisionof, 21 f.

Philostratus, 215.

<pv<ris, as title of early philosophicwritings, 29 f.; as nature, Xeno-phanes, 34; as origin, or primalsubstance, 47 ff.; opposed to Beats,

74 ff.; harmonized with vopos withStoics, 172, 209; Plotinus, 246.

Phurnutus, see Cornutus.Physico-theology, with Stoics, 195-

197; see Teleology.Pierre d'Ailly, 315, 333, 345.

Pittacus, 24.

Pity, see Sympathy.Plato, as authority for Socrates, 71,

77, 97; as systematiser, 99; generalcharacter of philos., 101; life andwritings, 102 f., grounds meta-physics anew, 105-109; Ideas, 116ff. ; doctrine of recollection, 118;of soul, 685; logic and dialectic, 119f.; Idea of Good, 122; his psychol-ogy, 123 f.; ethics, 125; politics,

126 f.; on education, 127; teleology,

128; doctrine of space, 129; im-portance of mathematics for, 129;philos. of Nature, 129 f.; relation

to Aristotle, 133, 139 ff.; on free-

dom, 191 ; influence of his dualism,211; regarded as starting-point fornatural science, 303.

Platonism, as a characteristic of Alex-andrian philosophy, 212; see alsoAcademy and Neo-Platonism.

Pleasure and Pain, referred to differ-

ences in motion, 86; as ethicalcriterian, 165, 170; see also Eudse-monism, Hedonism, Utilitarianism.

Pleroma, of Gnostics, 239.

Plotinus, 214 f., 218, 228, 233 ff., 237f., 244 ff., 290, 335.

Plurality, of substances, 39; deniedby the Eleatics, 37 f., 44; of co-

existing worlds, in Atomism, 54.

Plutarch of Chasr, 175, 213, 216, 221,

231 f., 225, 239.

Plutarch of Athens, 215, 234.

Pneuma, Stoic doctrine of, 186 f.

Polemo, 103.

Politics, see State.

Polus, 75.

Polybius, 176.

Porphyry, 31, 215, 218, 250 f., 288.

Posidonius, 161 f., 230.

Positivism, of Epicureans, 205.

Possibility, with Aristotle, 140; cf.

Potential, Actual, Necessity.Potential, 140, 144, 146.

Power, with Aristotle, 140; see alsoPotential, Force.

Pragmatic factor, in history of phi-losophy, 11-13.

Predestination, with Augustine, 284f

.; maintained by Thomas, and re-

jected by Scotus, 334.

Principium individualionis, 337, 341.

Principle of Contradiction, Zeno, 61

;

Protagoras, 88; Aristotle, 138.

Principle of Identity, Sophists, 89 f.

Probabilism, with Carneades, 207.

Proclus, 215, 218, 220, 222, 226, 228,238. 250 f.

Prodicus, 69, 71, 73, 76, 88, 96.

Protagoras, life, 70; ethical and reli-

gious views, 74, 76 ;perception the-

ory, 86, 91 f.; influence of this onDemocritus and Plato, 104 f., 117;relativism, 92, 105, 117; cf. 60 note1, 69, 88.

Psellos, see Michael.i'vx'h, see Soul.

Psychology, at first materialistic, 65

;

advanced by the Sophists, 69; ofthe Cyrenaics, 86; of Protagoras,91 ff.; of Democritus, 113-115; ofPlato, 123 f.; of Aristotle, 149 f.;

of Stoics, 168. 187 f.; 202-204; ofEpicureans, 202; of Plutarch, Ori-gen, etc., 232; of Neo-Platonism,234; of Augustine, 280-283 ; studiedin the Middle Ages, 303 ff. ; asso-

ciational, founded by John ofSalisbury, 307 ; metaphysicalpsych, of Thomas, Scotus, and Oc-cam, 324 f.; empirical phych. of

later Scholastics, 344 f.

Pyrrho, 160, 163, 165 ff., 200.

Pythagoras, 24, 30 f., 215.

Pythagoreans, 29 ff., 45 ff., 56 f., 60 f.,

72, 106 f., 120, 131, 147, 212 f.; see

also Neo-Pythagoreans.Pythagoreanism, 215.

Qualities, primary vs. secondary, 117;

all qualitative reduced to quantita-

tive by Democritus, 111; this op-posed by Aristotle, 148.

Quantitative, the only determina-tions recognized by Democritus,111.

Rabanus Maurus, 273.

Ramundua Lullus, see Lullus.

Ramundus of Sabunde, 315. 322.

Rationalism, of Pythagoreans, 45 f.;

of Cosmologists, 60 ; of Plato and

Page 388: Windelband, History of Philosophy, Vol. 1

Index.

Democritus, 105, 108, 110; of the

Stoics, 207 ff.; of Abelard, 300.

Realism, mediseval, 271 f.; of JohnScotus, 289; tends to pantheism,

295; modified, 297; criticized byAbelard, 298; of Scotus, 341.

Reality, grades of, 106, 250 f., 291 ff.

Reason, as motive-matter with An-axagoras, 41 f.; active and passive

with Aristotle, 150; Stoic doctrine

of, 171 f., 175, 180, 187 f., 223; op-posed to revelation by Tatian,

Tertullian, and others, 224 f.;

Philo's doctrine, 241 f.; and faith

with Abelard, 300 f.; with Albert

and Thomas, 321 f.; with Scotus

and Occam, 322 f.; active andpassive with Averroes, 339.

Recollection, Plato's doctrine of,

118 f.

Relativity of knowledge, Protago-rean, 92 f.; of Aristippus, 93 f.;

with Sceptics, 200 ff.; see also

Knowledge and Epistemology.Religion, relation to philos., in early

thought, 27; see Revelation, Chris-

tianity.

Renaissance, foreshadowing of, 302,

307.

Responsibility, 172, 192-194; presup-poses contingency of the will, 330f.; cf. Freedom.

Revelation, as tradition or ecstasy,

219 ff.; in relation to history, 223,

256 ff.; in relation to reason, 219

ff.; as equivalent to reason, 223 f.,

487; as opposed to reason, 224 f.,

322 f.

Richard of St. Victor, 275, 305.

Richard of Middletown, 314. 331, 333.

Right or law, philos. of, with Soph-ists, 74 f.; Socrates, 80 f.; Stoics

and Cicero, 177; Thomas, 326.

Rights, claim of equal, by Sophists,

74.

Robert Pullevn, 275.

Roscellinus, 274, 296 f., 298.

Saadjah Fa.ijumi, 318.

Sallustius, 218.

Salvation, by absorption into theOne, 250; by grace, 285; as centreof history, 256, 261 f.; as starting-

point for psychology, 305.

Saturninus. 214, 217, 239, 258.

Satyrus, 161.

Sceptics and Scepticism, ancient, 160,

163, 165, 170. 200. 205 f.

Scholastic method, 312 f.

Scholastics, -cism, 229. 266 ff.

Schools of philosophv, as associa-

tions, 6 f., 66, 70, 100'f., 103, 159 ff.;

see also Academy, Epicurus. Elean-

Eretrian, Socratic, Stoic, Peripa-

tetic.

Science, as equivalent to philosophy,

2 ; created by Greeks, 23 ; its essen-

tial nature, 95; see Philosophy,Natural Science.

Scotism, see Duns Scotus.

Scotus Erigena, see Erigena.Self-consciousness, Aristotle, 145

;

Neo-Platonists, 234; see Con-sciousness.

Self-enjoyment of personality, 116,

170.

Self-realisation, Aristotle, 151.

Seneca, 213, 215, 230.

Sensation, view of Cosmologists,

64 f.

; of Sophists, 91 ; of Democri-tus, 112 f.; of Aristotle, 149 f.;

Sensualism, Perception, Psychol-ogy.

Senses, world of, as a mixture of

Being and Non-being, with Plato,

129, with Plotinus, 247; as evil, 229

ff.; see also Sensation.Sensualism, of Protagoras, 91; of

Cynics, 96; of Stoics and Epicu-reans, 202 f.; basis for orthodoxy,225 ; as a consequence of Nominal-ism, 297.

Sermonism, 272, 298.

Seven Wise Men, 24.

Sextians, 161, 163.

Sextus Empiricus, 160, 163.

Signs, theory of, 325 f., 343.

Simon of Tournay, 320.

Simplicius, 31 f., 219, 339.

Sin, as error, with Socrates, 80, 191;

as a falling away, 253 ; as problemof theodicy, 197.

Society, utilitarian theory of, amongEpicureans, 173 f.; need of, 328.

Socrates, general character of his

teaching, 69 f.; life, 71; ethical doc-trine, 76-82; theory of knowledge,94-98; influence on Plato, 101 f.,

107 f., 116, 118 f.; doctrine of free-

dom, 191 ; cf . also Socratic Schools.

Socratic Schools, 70 ff., 82 ff., 89, 96.

Solon, 24, 34.

Sophists, 25 f., 67 ff., 73 ff., 88 ff., 221.

Sophocles, 74.

Sotion, 161, 163, 215.

Soul (see also Psychology, Self) first

conceived as moving force, 62 ff.;

with Democritus, consists of atoms,

113; pre-existence of, 119, 123, 230

f., 249; transmigration of, 62, 119,

232; its twofold aspect with Plato,

123 f.; immortality of, with Plato,

124; three souls with Aristotle, 149

f.; Stoic view of, 187 f.; Epicurean,

Page 389: Windelband, History of Philosophy, Vol. 1

Index.

188; conception analysed by Alex-andrian thought, 232; contrasted

with spirit and vital force, 232 f.;

divided into higher anil lower byPlotinus, 246 ; unity of, withAugustine, 278; faculties of, see

Psvchology; see also Immortality.Soul' and body, 301 ff.

Space, with Parmenides= Non-being,37; exists, Leucippus, 42; withPlato= Non-being, 129; accessory

cause of world, 129, 131.

Species (logical), with Plato, see

Ideas; with Aristotle, 142; in Mid-dle Ages, see Universals; intelli-

qibiles, 325, 343.

Speusippus, 103, 123, 164, 243.

Spirit, not equivalent to immaterialwith Plato, 118; pure Form withAristotle, 145; opposed to matterby later Stoics, 230; by Apologists,

231 f.; by Alexandrian thought,231-235; made equivalent to im-material, 229; contrasted with soul,

232 f.

Spirits, animal, 187.

Spiritualisation, of the universe, 249,

253.

State: Cynic attitude toward, 84 f.;

Cyrenaic, 86; Plato's doctrine of,

126; Aristotle's, 152 f.; Epicureandoctrine of, 173 f.; Stoics, 173-177;Thomas on, 326 f.; Dante, 327; Oc-cam, 328; see also Contract theory.

Stoics and Stoicism, general, 157, 159

;

personality and writings, 162;

ethics, 164 ff., 167 ff., 171 ff.; viewof society, 175; view of Nature, 180

ff. ; conception of law and provi-dence, 180 f

.; theory of knowledge,

202 ff., 207 ff.; cf. also 210 f., 221,

223, 230.

Strato, 159, 161, 179, 180, 199, 338.

Strife, as principle, Heraclitus, 50.

Subjective=real, 325; subjectivity

of sense-perception, 60 note, 92,

105. 112 f.

Sublime, Longinus on, 218Subordination, as the essential char-

acteristic of judgment, 135.

Substance, two elements in first con-ception of, 35 f.; Aristotle's con-ception, 143; Stoic, matter, 199;

category of, not applicable to God,Augustine, 279 f.

<TvyKo.Ta6ecns, 207, 308.

Summists, 275, 313.

Supersensuous, 117, 323; cf. World,Reason.

Superstition, among the Stoics, 189 f.

Supposition, 325 f., 342 f.

Survival of the fittest, 53, 185.

Suso. 314.

Suspense, of judgment, among Scep-tics, 167, 202.

Syllogism, Aristotle's doctrine of,

135; criticized by Sceptics, 201.

Symbolism, of numbers, see Num-bers.

Sympathy, in tragedy, 153.

Syncretism, 161.

Svnergism, 334.

Synteresis, 333.

Svnthesis, Neo-Platonist doctrine of,

234.

Syrianus, 218.

Syzygies, of the Gnostics, 244.

Tatian, 214, 217, 224.

Tauler, 314.

Teleology, of Anaxagoras, 42, 54; ofSocrates, 98; Plato, 128; Aristotle,

144 ff., 178; rejected by Strato, 179;maintained by Stoics, 181 f., 195 ff.;

opposed by Epicureans, 182 f.

; ofhistory, among Christian thinkers,

260 ff.

; of Nature and history withThomas, 327.

Teles, 216.

Terminism, of Occam, 325, 342; seealso Nominalism.

Tertullian, 214, 217, 221 f., 224 f.

Thaies, 24, 27, 29, 32 f., 48 f.

Themistius, 218.

Theodicv, of Stoics, 196 f.; of Ploti-

nus, 247; Patristic, 252 f.; of Au-gustine, 280, 283 f.; see also Tele-ology, and Evil.

Theodoric of Chartres, 294, 302.

Theodorus, 70, 86 f.

Theogomy, of the Gnostics, 243 f.

Theology, combination with philos-ophy, Pythagoreans, Plato, 62;Aristotle's, 145 f.; natural and re-

vealed, 321 f.; theol., separatedfrom philosophy aa "practical," bvScotus, 333.

Theophilus, 217.

Theophrastus, 103, 159, 161, 164, 178f., 198.

öecopia, 154, 250, 286, 333.

Theurgy, 250.

Thomas Aquinas and Thomism, as

systematiser, 311 ; life and writings,

313; on faith and reason, 321; psy-chology, 324; doctrine of the state,

326 f.

; on relation between will andintellect, 328 ff.; determinism, 330;God and the good, 332; ethics. 333;on problem of individuality, 340 f.:

opposed by empiricism, 344 f. Cf.

also 299.

Thought, opposed to perception byCosmologists, 58 ff.; by Plato andDemocritus, 105; as object of

Page 390: Windelband, History of Philosophy, Vol. 1

Index.

Logic, 133; as the divine self-con-

sciousness, with Aristotle, 145; see

also Reason, and Epistemology.Thrasyllus, 162, 216.

Thrasymachus, 75 f.

Tiedemann, 10.

Timon, 160, 167, 200.

Tragedy, defined by Aristotle. 153.

Transcendence, of God, 146, 236, 338;cf. Immanence.

Trinity, doctrine of Augustine, 2S0;in Realism, 295; above reason. 321.

Tropes of ^Enesidemus, 200; of Ag-rippa, 201.

Truth (see Epistemologv), twofold,

320 ff.

Twofold truth, doctrine of, 320 ff.

Ueberweg, 16.

Uniformity of Nature, understood in

Astronomy, 57; taught bv Demo-critus, 107 ff. and the Stoics. ISO f.,

194 f.; restricted by Aristotle, 142f.; denied by Epicurus, 182 f.; seealso Law and Nature.

Unity, the Eleatics, 38; Spinoza, of

consciousness, 150 ; of human race,

261 ; see also One.Universal and particular (see also

Realism. Nominalism, and Con-ceptionalism), Socrates, 97; Plato's

Ideas as. 107-109, 119-122; Aris-

totle, 133 ff ., 142 f. ; mediaeval con-troversy over, 287 ff.; Abelard, 299.

LTtilitarianism, with Sophists andSocrates, 74 f., 78 f.; with Epi-curus, 174 f.

Utopia, 126.

Valentinus, 214, 217, 239, 243, 254.

Validity, universal, as problem of theSophists, 68, 74, 93; postulated bySocrates, 69, 81, 95 ff.

Values, conception of, introducedinto theoretical consideration byAnaxagoras, 42; by Anaximander,49; by Pythagoreans, 57; by Platoand Democritus, 106; by Aristotle,

143; measured in terms of pleas-ure, see Hedonism; by otherstandards, see Good.

Varro. 161. 163.

/ictorines, 275. 305. 323.Vincent of Beauvais. 313. 344.Virtue, ambiguity of the term, 78;

consists in knowledge. Socrates, 7Sff.; necessarily results in happiness,SI; the sole good, Antisthenes. 83;is ability for enjoyment, Aristip-pus, 85; is knowledge. 184; is sus-pense of judgment. Sceptics, 167;is sole good for Stoics, 16S; arises

only through the logos, Philo, 227;as stimulated by beauty, Plotinus,

250; see also Ethics and Good.Virtues, the four cardinal, of Plato,

125 f.; ethical and dianoetic, of

Aristotle, 151, 154; these subordi-nated to the Christian by Augus-tine. 2S7; dianoetic. above thepractical, with Thomas, 333; cf.

Ethics.

Void (see also Space), with Parme-nides, 3S; with Gnostics. 239.

Voluntarism, of Augustine, 281 ff.;

of Scotus, 328 ff.

Walter of Montagne, 274.

Way upward and downward, Hera-clitus, 36, 50.

Will (see also Freedom) with Au-gustine. 2S1 f.; relation to intel-

lect, 328 ff.; see Voluntarism.William of Auvergne, 314, 332. 340.

William of Champeaux, 272, 274, 294,298.

William of Conches, 274, 302 f

.

William Durandus, 315.

William de la Marre, 314.

William of Occam, 312, 315, 322 f.,

325, 328, 330, 342 ff.

Wise man, Stoic ideal of, 169, 171 ff.;

Epicurean, 166, 170 f.

Wise Men, the Seven, 24.

World, early Greek views of, 31 ff.;

of Pythagoreans, 47, 57; of Plato,123; of Middle Ages, 294, 323;origin of, according to early Greeks,48 ff.; plurality of, 49; a process.Heraclitus, 50; incorporeal, ofPlato, 117 f.; origin of, with Plato,

130; Aristotle's view of. 147 f.; Epi-curus, 184; as macrocosm. Stoics,

1S7; product of spirit. 235; rela-

tion of to God, 235 ff. ; eternal, ofOrigen, 254; intelligible, 290. Seealso Nature, Natural Science, His-torv.

World-reason, 128, 172, 187. See alsoLogos.

World-soul. 63 note 1, 131 ; evil withPlato. 231.

World-stuff, 32 ; as water, as air, 32,

55; as fire, 36.

Neniades. S4.

Xenocrates, 103. 123, 164. 243.Xenophanes. 28, 30. 34 ff.. 46. 146.Xenophon, 71, 77 ff., 82, 97, 1S2.

Zeno of Elea, 28. 30, 44, 55 f., 61,89 ff.

Zeno of Sidon, 162.

Zeno the Stoic, 159, 162, 168, 175.

Page 391: Windelband, History of Philosophy, Vol. 1

HARPER TORCHBOOKS / The Cloister Library

Tor AndraeAugustine/PrzywaraRoland H. Bainton

Karl BarthKarl Barth

Nicolas BerdyaevNicolas Berdyaev

James Henry Breasted

Martin BuberMartin Buber

Jacob Burckhardt

Edward ConzeF. M. Cornford

G. G. CoultonG. G. Coulton

H. G. CreelAdolf Deissmann

C. H. DoddJohannes Eckhart

Mircea Eliade

Morton S. EnslinMorton S. Enslin

Austin Farrer, ed.

G. P. Fedotov

Ludwig FeuerbachHarry Emerson Fosdick

Sigmund Freud

Maurice FriedmanOctavius Brooks Frothingham

Edward Gibbon

Edward Gibbon

Charles C. Gillispie

Maurice Goguel

Maurice Goguel

Edgar J. GoodspeedHerbert J. C. Grierson

William HallerAdolf HarnackEdwin HatchKarl Heim

F. H. HeinemannStanley k. Hopper, ed.

Johan HuizingaAldous HuxleyImmanuel Kant

Sdren KierkegaardSdren KierkegaardS0ren KierkegaardAlexandre Koyre

Emile Male

T. T. MeekH. Richard NiebuhrH. Richard Niebuhr

H. J. RoseJosiah Royce

mohammed : The Man and His Faith tb/62AN AUGUSTINE SYNTHESIS TB/35THE TRAVAIL OF RELIGIOUS LIBERTY TB/3ODOGMATICS IN OUTLINE TB/56THE WORD OF GOD AND THE WORD OF MAN TB/13THE BEGINNING AND THE END TB/14THE DESTINY OF MAN Tb/6iDEVELOPMENT OF RELIGION AND THOUGHT IN ANCIENT

EGYPT TB/57ECLIPSE OF GOD TB/l2moses : The Revelation and the Covenant TB/27THE CIVILIZATION OF THE RENAISSANCE IN ITALY [Illus-

trated Edition] : Vol. I, TB/40 ; Vol. II, T3/41buddhism : Its Essence and Development TB/58from religion to philosophy : A Study in the Origins

of Western Speculation TB/20MEDIEVAL FAITH AND SYMBOLISM TB/25THE FATE OF MEDIEVAL ART IN THE RENAISSANCE AND

REFORMATION TB/26CONFUCIUS AND THE CHINESE WAY TB '63

paul: A Study in Social and Religious History TB/15THE AUTHORITY OF THE BIBLE TB/43meister eckhart : A Modern Translation tb/8cosmos and history : The Myth of the Eternal Return

TB./50CHRISTLAN BEGINNINGS TB/sTHE LITERATURE OF THE CHRISTIAN MOVEMENT TB/6THE CORE OF THE BIBLE Tb/7the Russian religious mind : Kievan Christianity, the

10th to the 13th Centuries tb/"oTHE ESSENCE OF CHRISTIANITY Tb/iIa guide to understanding the bible tb/2on creativity and the unconscious : Papers on the

Psychology of Art, Literature, Love, Religion tb .'45

martin buber : The Life of Dialogue TB/64TRANSCENDENTALISM IN NEW ENGLAND TB 59THE END OF THE ROMAN EMPIRE IN THE WEST [J. B. Bury

Edition ; Illustrated] TB/37THE TRIUMPH OF CHRISTENDOM IN THE ROMAN EMPIRE

[J. B. Bury Edition; Illustrated] TB/46GENESIS AND GEOLOGY TB/51jesus and the origins of Christianity i : Prolegom-

ena to the Life of Jesus TB./65

jesus and the origins of Christianity ii : The Lifeof Jesus tb/66

A LIFE OF JESUS TB/lCROSS-CURRENTS IN I7TH CENTURY ENGLISH LITERATURE:

The World, the Flesh, the Spirit TB/47THE RISE OF PURITANISM Tß/22WHAT IS CHRISTIANITY? TB/17THE INFLUENCE OF GREEK IDEAS ON CHRISTIANITY TB 'lS

CHRISTIAN FAITH AND NATURAL SCIENCE TB,'l6EXISTENTIALISM AND THE MODERN PREDICAMENT TB '2SSPIRITUAL PROBLEMS IN CONTEMPORARY LITERATURE TB 21ERASMUS AND THE AGE OF REFORMATION TB IQTHE DEVILS OF LOUDUN TB ,'6o

RELIGION WITHIN THE LIMITS OF REASON ALONE TB/67edifying discourses: A Selection tb '32THE JOURNALS OF KIERKEGAARD TB/52PURITY OF HEART TB/4FROM THE CLOSED WORLD TO THE INFINITE UNIVERSE TB 3 I

the gothic image: Religious Art in France of the 13thCentury TB/44

HEBREW ORIGINS TB/60CHRIST AND CULTURE TB/3THE KINGDOM OF GOD IN AMERICA TB/49RELIGION IN GREECE AND ROME TB/55THE RELIGIOUS ASPECT OF PHILOSOPHY TB/20

(.Continued on next page)

Page 392: Windelband, History of Philosophy, Vol. 1

Auguste Sabatier

George SantayanaGeorge Santayana

Friedrich SchleiermacherHenry Osborn Taylor

Paul Tillich

Edward Burnett TylorEdward Burnett Tylor

Evelyn Underhill

Evelyn UnderhillJohannes Weiss

Wilhelm Windelband

Wilhelm Windelband

OUTLINES OF A PHILOSOPHY OF RELIGION BASED ON PSY-

CHOLOGY AND HISTORY TB/23INTERPRETATIONS OF POETRY AND RELIGION Tb/qwinds of doctrine and

PLATONISM AND THE SPIRITUAL LIFE TB/24on religion : Speeches to Its Cultured Despisers TB/36THE EMERGENCE OF CHRISTIAN CULTURE IN THE WEST \

The Classical Heritage of the Middle Ages TB/48DYNAMICS OF FAITH TB/42THE ORIGINS OF CULTURE TB/33RELIGION IN PRIMITIVE CULTURE TB/34the golden sequence : A Fourfold Study of the Spir-

itual Life tb/68worship tb/ioearliest Christianity : A History of the Period a.d.

50-750: Vol. I, TB/53 ; Vol. II, TB/54a history of philosophy I : Greek, Roman, Medieval

TB/38a history of philosophy 11 : Renaissance, Enlighten-

ment, Modern TB/39

HARPER TORCHBOOKS / The Academy Library

H. J. BlackhamWalter Bromberg

Editors of Fortune

G. P. Gooch

Francis J. GrundW. K. C. Guthrie

Henry JamesHenry JamesHenry JamesArnold Kettle

Arthur O. Lovejoy

J. P. Mayer

Samuel Pepys

Georges PouletFerdinand Schevill

W. Lloyd Warner

six existentialist thinkers TB/1002the mind of man : A History of Psychotherapy and

Psychoanalysis TB/1003America in the sixties : The Economy and the Society

TB/1015ENGLISH DEMOCRATIC IDEAS IN THE SEVENTEENTH CEN-

TURY TB/1006ARISTOCRACY IN AMERICA TB/lOOITHE GREEK PHILOSOPHERS TB/1O08THE PRINCESS CASAMASSIMA TB/lO05RODERICK HUDSON TB/lOl6THE TRAGIC MUSE TB/1017AN INTRODUCTION TO THE ENGLISH NOVEL \ Vol. I, Defoe

to George Eliot, tb/ioii; Vol. II, Henry Jamesto the Present, TB/1012

the great chain of being : A Study of the History of anIdea tb/iooq

Alexis de tocqueville : A Biographical Study in Political

Science TB/1014the diary of samuel pepys: Selections, ed. by O. F.

Morshead ; illus. by Ernest H. Shepard TB/1007STUDIES IN HUMAN TIME TB/1OO4THE MEDICI TB/lOIOSOCIAL CLASS IN AMERICA TB/1II3

HARPER TORCHBOOKS / The Science Library

R. B. BraithwaiteLouis de Broglie

J. BronowskiT. G. Cowling

W. C. Dampier, ed.

Arthur Eddington

H. T. Pledge

George SartonPaul A. Schilpp, ed.

O. G. SuttonStephen Toulmin

Andrew G. Van MeisenFriedrich Waismann

W. H. Watson

G. J. Whitrow

A. Wolf

SCIENTIFIC EXPLANATION TB/515PHYSICS AND MICROPHYSICS TB/514SCIENCE AND HUMAN VALUES TB/505MOLECULES IN MOTION TB/516READINGS IN THE LITERATURE OF SCIENCE TB/SI2space, time and gravitation : An Outline of the Gen-

eral Relativity Theory tb/sioscience since 1500: A Short History of Mathematics,

Physics, Chemistry, and Biology tb/so6ANCIENT SCIENCE AND MODERN CIVILIZATION Tb/SOIalbert einstein: Philosopher-Scientist : Vol. I,

TB/502; Vol. II, TB/503MATHEMATICS IN ACTION TB/518THE PHILOSOPHY OF SCIENCE TB/513FROM ATOMOS TO ATOM TB/517INTRODUCTION TO MATHEMATICAL THINKING TB/51Ion understanding physics : An Analysis of the Philos-

ophy of Physics TB/507THE STRUCTURE AND EVOLUTION OF THE UNIVERSE : A»

Introduction to Cosmology TB/504A HISTORY OF SCIENCE, TECHNOLOGY AND PHILOSOPHY IN

THE 1 6i'H AND 17TH CENTURIES: Vol. I, TB/508;

Vol. II, TB/509

Page 393: Windelband, History of Philosophy, Vol. 1
Page 394: Windelband, History of Philosophy, Vol. 1
Page 395: Windelband, History of Philosophy, Vol. 1
Page 396: Windelband, History of Philosophy, Vol. 1

DATE DUEDUE RETURNED

nUU U uUUJMäR 2 8 E0Ö4

A fine of 25 cents per day is charged when this item is overdue.

Page 397: Windelband, History of Philosophy, Vol. 1

UNIVERSITY OF FLORIDA

3IIIlIll

3 1262 04824 8602

Page 398: Windelband, History of Philosophy, Vol. 1